0% found this document useful (0 votes)

279 views

Citi Quantum Computing

This document discusses the potential of quantum computing and its advantages over classical computing. Quantum computing uses quantum mechanics and can potentially solve certain problems much faster than classical computers. While still theoretical, quantum computing is advancing quickly and may revolutionize industries like logistics, drug discovery, finance and more. Nations and companies should begin preparing for its arrival through funding research, education, and impact assessments to understand how it could transform their industries. The total addressable market for quantum computing is estimated to reach between $700 million to $8.6 billion by 2027.

Uploaded by

goondu_93

Available Formats

Download as PDF, TXT or read online on Scribd

0% found this document useful (0 votes)

279 views

Citi Quantum Computing

Uploaded by

goondu_93

Available Formats

Download as PDF, TXT or read online on Scribd

You are on page 1/ 152

QUANTUM COMPUTING

Moving Quickly From Theory to Reality

Citi GPS: Global Perspectives & Solutions

July 2023

Citi is one of the world’s largest financial institutions, operating in all major established and emerging markets. Across these world markets, our employees conduct
an ongoing multi-disciplinary conversation – accessing information, analyzing data, developing insights, and formulating advice. As our premier thought leadership
product, Citi GPS is designed to help our readers navigate the global economy’s most demanding challenges and to anticipate future themes and trends in a fast-changing and
interconnected world. Citi GPS accesses the best elements of our global conversation and harvests the thought leadership of a wide range of senior professionals
across our firm. This is not a research report and does not constitute advice on investments or a solicitations to buy or sell any financial instruments.
For more information on Citi GPS, please visit our website at www.citi.com/citigps.
Citi GPS: Global Perspectives & Solutions July 2023

Authors
Tahmid Quddus Islam Robert Garlick Wenyan Fei
Analytical Insights Team Head of Innovation, Analytical Insights Team
Citi Global Insights Technology, and the Citi Global Insights
Future of Work
+44-20-7986-3503 | Citi Global Insights +44-20-7986-3543 |
tahmid.quddus.islam@citi. [email protected]
com +44-20-7986-3547 |
[email protected]
Professor Deeph Dr. Stefano Gogioso Professor Sougato
Chana Computer Scientist & Bose
Chair of the NATO Researcher in Quantum Physicist and
Advisory Group on Computing Researcher in Quantum
Emerging and Oxford University Computation
Disruptive Technologies University College
London

Sean Kornish David W. Edelman Anuj Gangahar

Chair Co-chair Analytical Insights Team
Citi Cryptography Citi Cryptography Citi Global Insights
Security Center of Security Center of
Excellence Excellence [email protected]

[email protected] [email protected]

Helen H. Krause, CFA Yehuda Dayan Devon Stone

Head of Data Science Data Scientist Data Scientist
Insights Citi Global Insights Citi Global Insights
Citi Global Data Insights
+44-20-7986-5502 | +44-20-7986-8691 |
+44-20-7986-8653 | [email protected] [email protected]
[email protected]

Ronit Ghose, CFA Kaiwan Master Atif Malik

Head of Future of Future of Finance Semiconductor Cap
Finance Analyst Equipment & Specialty
Citi Global Insights Citi Global Insights Semiconductor Analyst
Citi Research
[email protected] kaiwan.hoshang.master@c
iti.com +1-415-951-1892 |
[email protected]
Ronald Josey Carol Gibson
U.S. Internet Analyst Analytical Insights Team
Citi Research Citi Global Insights

+1-212-816-4545 | +44-20-7986-4137 |
[email protected] [email protected]
July 2023 Citi GPS: Global Perspectives & Solutions 3

QUANTUM COMPUTING
Moving Quickly From Theory to Reality
Kathleen Boyle, CFA Technology advances seem to be dominating the headlines. There are new ways to
Managing Editor, Citi GPS search the web using artificial intelligence and machine learning, alternate payment
systems using blockchain and digital currency, and innovative ways to socialize
gaming, and shop using the metaverse. In most cases, the advances have been
facilitated by increasing computing speeds that enable faster calculations as well as
better connectivity, but all are based on existing computing technology.

But that is all about to change. The next phase of computing — quantum computing
— is getting close to moving from being theoretical to practical, and we are quickly
coming to a point where quantum computers will be able to perform tasks faster,
more efficiently, and cheaper than classical computers. Quantum computing breaks
away from traditional classical computing using quantum mechanics as a base. If
you made it through physics class in high school or college, you might remember
that quantum mechanics as a daunting concept associated with geniuses like Albert
Einstein, Niels Bohr and Max Planck. But you don’t have to be a rocket scientist to
start thinking about how quantum computing can revolutionize industry and society.
And given the speed that quantum computing is progressing, we believe now is the
time to for nation-states, corporates, and market participants to start getting ready
for its arrival.

In the report that follows, we look at what quantum computing is and what its
advantages are compared with classical computing in terms of optimization,
machine learning, simulation, and cryptography. We then apply those advantages to
specific industries to highlight areas where quantum computing can materially
advance processes from logistics and drug discovery to portfolio optimization and
cybersecurity.

Most importantly, we set out steps that nation-states and corporates can begin
implementing in preparation for the arrival of quantum computing. Setting up
appropriate funding to encourage research and development is important on a
national scale, as is ensuring education and training to avoid a talent skills gap.
Corporates need to improve their awareness of how quantum computing will affect
their industry, create impact assessments, and contextualize opportunities.

© 2023 Citigroup
THE QUANTUM COMPUTING OPPORTUNITY
WHY QUANTUM COMPUTING IS A BIG DEAL
Computing is at an inflection point. After five
Moore’s Law of Exponential Growth vs.
decades of increasing computational power Neven’s Law of Doubly Exponential Growth
by Moore’s Law (i.e., doubling the number of
transistors per chip every 2 years), the growth

Computing Power, Log Scale

rate of classical computing is reaching its physical
limit. At the same time, the capabilities of quantum
computers (QCs) are advancing at a “doubly
exponential” rate — in line with Neven’s Law —
and are close to generating commercial value.

Time
Exponential Growth Doubly Exponential Growth

HOW BIG COULD QUANTUM COMPUTING BECOME?

Total addressable market (TAM) estimates for quantum computing vary widely — ranging in 2027 from
$700 million to $8.6 billion. Venture capital investment in the QC space has been strong but still pales in
comparison to investment in artificial intelligence (AI).

Range of Current TAM Forecasts (2022-27E) VC Investment in Quantum Computing (2010-22)

Source: Industry forecasts Source: PitchBook Data Inc.
($bn) ($bn)
10
2.0
9
8
1.5
7
6
1.0
5
4
0.5
3
2
1 0.0

0 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022

2022 2023 2024 2025 2026 2027 Angel & Seed Early-stage VC Later-stage VC

VC Investment in Quantum Computing vs. Artificial Intelligence

Source: PitchBook Data Inc.
($bn)

120

100

0
2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022

Quantum Computers Artificial Intelligence

HOW WILL QUANTUM COMPUTING DISRUPT INDUSTRY?
Quantum computers will likely solve practical problems faster, cheaper, or more efficiently, than classical
computers in four broad areas — optimization, machine learning, simulations, and cryptography. Three of
these attributes offer significant upside for numerous industries, including Manufacturing & Logistics,
Artificial Intelligence, Healthcare, Energy & Climate, and Finance. However, with cryptography, QCs pose a
potential threat to existing cryptographic standards that underpin computer systems and cryptocurrencies.

MANUFACTURING ARTIFICIAL ENERGY

AND LOGISTICS INTELLIGENCE AND CLIMATE

Transition to Industry 5.0 Enhanced natural Improved crude

language processing refinery processes
Advanced materials discovery
Artificial neural networking Economically viable
Enhanced digital twin processes
green hydrogen
Logistics optimization
Accelerated battery
technology development
FINANCE HEALTHCARE
Ammonia synthetization
More personalized services Expedited drug
discovery and delivery New catalysts for
Enhanced portfolio optimization
carbon capture
Improved risk management/ Improved diagnoses
fraud detection Tailored treatments
Better insurance risk
assessment/fraud detection

PREPARING FOR QUANTUM

Due to the exponential scaling power of quantum computers, by the time they offer a practical advantage to
business, the gap between early adopters and those using only classical computing will widen at an increasingly
accelerated pace. Now is the time to prepare for the transition to capitalize on opportunities and safeguard
against risk. To take advantage of the opportunity…

GOVERNMENTS SHOULD: CORPORATES SHOULD:

Establish a holistic quantum policy Build quantum awareness

Invest in quantum infrastructure Develop preparedness
Upskill workforce Position yourself for quantum advantage through
investigation and technology investment
Develop international
strategic initiatives Understand and develop an enterprise approach
to cryptography modernization and agility
Consider future ethical concerns

Contents
Executive Summary 8
Quantum Computing 101 13
Introduction ................................................................................................. 13
Why Quantum Computing May Be the Defining Technology of the 2020s 14
Understanding Quantum Computers .......................................................... 17
The Rapid Development of Quantum Computers ....................................... 20
Areas Quantum Advantage Could Occur.................................................... 22
Expert Interview with Nobel Laureate, Professor David J. Wineland, at
the University of Oregon ....................................................................... 24
A Hybrid Approach to Computing Is Inevitable ........................................... 26
Expert Interview with Physicist, Professor Sougato Bose, at University
College London .................................................................................... 28
The State of the Quantum Computing Market ............................................ 29
Expert Interview with Dr. Dario Gil, Senior Vice President and Director
of Research at IBM Quantum ............................................................... 34
Industry Impact 36
Manufacturing and Logistics ....................................................................... 37
Material Discovery ................................................................................ 37
Production Processes ........................................................................... 38
Supply Chains ...................................................................................... 38
Expert Interview with Dr. Alan Baratz, CEO of D-Wave ....................... 40
Artificial Intelligence .................................................................................... 41
Natural Language Processing .............................................................. 42
Artificial Neural Networks ..................................................................... 42
Expert Interview with Jack Hidary, CEO of SandboxAQ ...................... 44
Healthcare ................................................................................................... 45
Drug Discovery ..................................................................................... 45
Medical Services .................................................................................. 45
Expert Interview with Peter Chapman, CEO of IonQ ........................... 48
Energy and Climate..................................................................................... 49
Energy .................................................................................................. 49
Agriculture............................................................................................. 50
Climate Change .................................................................................... 50
Expert Interview with Rajeeb Hazra, CEO of Quantinuum................... 52
Finance ....................................................................................................... 53
Targeting and Prediction ...................................................................... 53
Portfolio Optimization ........................................................................... 53
Risk Management and Fraud Detection ............................................... 54
Expert Interview with Matt Johnson, CEO of QC Ware ........................ 55
Cybersecurity .............................................................................................. 56
How Data Is Currently Protected .......................................................... 56
The Threat to Encryption Standards .................................................... 57
Post-Quantum Cryptography ................................................................ 59
Expert Interview with Dr. Michele Mosca, CEO and Co-Founder of
evolutionQ............................................................................................. 63
Cryptocurrency ............................................................................................ 64
Bitcoin’s Cryptography .......................................................................... 64
Cryptocurrency Mining.......................................................................... 65
The Broader Cryptocurrency Ecosystem ............................................. 66
Potential Courses of Action .................................................................. 67
Understanding the Landscape and How to Prepare 69
Nation-States .............................................................................................. 70

Government Investment ....................................................................... 70

Expert Interview with Roger McKinlay, Head of the Quantum
Technologies Challenge for UK Research and Innovation .................. 78
Workforce Education ............................................................................ 79
Expert Interview with Computer Scientist and Researcher, Dr. Stefano
Gogioso, at Oxford University............................................................... 86
Supply Chains ...................................................................................... 88
Expert Interview with Dr. Anthony J. Yu, Vice President of Silicon
Photonics Product Management at GlobalFoundries ........................... 91
Ethics .................................................................................................... 92
Expert Interview with Nick Farina, CEO of EeroQ ................................ 94
How to Prepare: A Holistic Quantum Computing Policy ...................... 95
Expert Interview with Celia Merzbacher, Executive Director of the U.S.
Quantum Economic Development Consortium (QED-C) ..................... 96
Corporates .................................................................................................. 97
Market Forecasts .................................................................................. 97
Expert Interview with Alex Challans, CEO of The Quantum Insider .. 101
Value Creation .................................................................................... 102
Expert Interview with William Hurley, CEO of Strangeworks ............. 106
Collaboration and the Cloud ............................................................... 107
Expert Interview with Professor Simone Severini, Director of Quantum
Computing at Amazon Web Services (AWS) ..................................... 109
How to Prepare: For Quantum Advantage ......................................... 111
Expert Interview with Dr. Christopher Savoie, CEO of Zapata
Computing .......................................................................................... 114
How to Prepare: For the Quantum Threat .......................................... 115
Expert Interview with Professor Deeph Chana, Chair of the NATO
Advisory Group on Emerging and Disruptive Technologies ............... 119
Market Participants ................................................................................... 120
Venture Capital Trends ....................................................................... 120
Expert Interview with Stuart Woods, Chief Operating Officer of
Quantum Exponential ......................................................................... 124
The Company and Funding Environment........................................... 125
Expert Interview with David Moehring, General Partner at Cambium
Capital................................................................................................. 128
How To Prepare: A Deeper Understanding........................................ 129
Expert Interview with Mark Danchak, Partner at General Innovation
Capital................................................................................................. 130
Closing Statement 131
Appendix: How Quantum Computers Work 132
The Basics of Quantum Physics ............................................................... 132
The Basics of Computing .......................................................................... 136
How Qubits Make Quantum Computers So Powerful............................... 138

Executive Summary
“Never mind AI (artificial intelligence) — quantum computing is the next big thing,”
said Martha Lane Fox, president of the British Chambers of Commerce, in June
2023.1 While this is up for debate as we marvel at the innovation taking place
currently in AI, quantum computing may become the defining technology of the
2020s. We believe quantum computing will radically change the way things are
done and commercial advantages to business will arrive sooner than many expect.
Both nation-states and corporates need to prepare. We hope this report helps raise
awareness and preparation as we move into an era of quantum computing.

State of the Quantum Computing Industry

Moore’s Law, the observation that transistor density in integrated circuits tends to
double every 18-24 months, has driven huge progression in computing but is
slowing down. The next leap for computing is quantum. Rather than the linear
scaling capabilities of our current computers (otherwise known as “classical
computers”), quantum computers (QCs), by their nature, scale exponentially.

Many believe we have already passed the point of “quantum supremacy,” the point
at which we can perform tasks with QCs beyond the capabilities of classical
computers. The next major milestone in the development of QCs is that of
“commercial quantum advantage” — i.e., the phase where QCs can solve practical
problems faster, more cheaply, or more efficiently than classical computers — and
is something most companies feel will come around the middle to the end of this
decade. While progress announced by IBM in June 2023 has increased optimism,
commercial quantum advantage will materialize at different times in different use
cases — and in distinctly granular ways.2 After the point of commercial quantum
advantage, we will have leapt into what many will consider the next stage of
computing.

QCs process information by harnessing the principles of quantum mechanics — the

laws that govern the behavior of atoms and photons. The basic building block of a
QC is a quantum bit or “qubit.” Qubits can simulate their classical counterparts —
occupying a state of either 0 or 1 — but can also be in a combination of both states
at the same time through the quantum mechanical property of “superposition.”
Furthermore, due to the process of quantum entanglement, every additional qubit
doubles the number of potential states a QC can be in. In 2019, the Director of
Google’s Quantum Artificial Intelligence Lab, Hartmut Neven, noted at a conference
that QCs were gaining computational power relative to classical ones at a “doubly
exponential” rate. His observation was eventually dubbed “Neven’s Law.” He went
on to describe the rate of progress as “It looks like nothing is happening, nothing is
happening, and then whoops, suddenly you’re in a different world.” If so, this would
follow the recent inflection point in AI, but we note a hybrid approach to computing is
inevitable, where QCs are used in tandem with classical computers.

QCs are highly specialized in nature and currently commercial quantum advantage
is expected in four specific areas:

1. Optimization: QCs are likely to improve on the ability to find the most efficient
answer to a problem with numerous potential solutions.

1 Martha Lane Fox, “I’ve Seen the Future and It Could Be Great for Britain,” The Times,
June 18, 2023.
2 Davide Castelvecchi, “IBM Quantum Computer Passes Calculation Milestone,” Nature,

June 14, 2023.

2. Machine Learning: QCs may accelerate existing workloads or unlock entirely

novel methods outright, impacting and further enhancing the development of
artificial intelligence.

3. Simulation: QCs are expected to have an exponential speed-up over classical

computers when it comes to the simulation of molecules and other quantum
systems.

4. Cryptography: QCs will be able to significantly weaken — or in some cases,

break — certain cryptographic standards.

Comparing different QCs is difficult given the existence of numerous qubit

technologies (e.g., superconducting, trapped ion, photonic, etc.), and widely varying
qubit quality depending on model and company.

Industry Use Cases and Risks

Manufacturing and logistics are widely considered to be among the main early
beneficiaries of quantum computing. In contrast to the limitations of classical
computers, QCs are expected to enable the comprehensive modeling of
sophisticated molecules, dramatically benefiting material discovery in the
manufacturing industry. QCs are expected to enhance the ability of digital twins
(virtual simulations of products or system lifecycles that can be updated from real-
time data) to monitor and automate production processes, as well as dramatically
lower costs. QCs are also expected to improve operational capabilities in industries
that require the consideration of a high number of complex variables such as
logistics systems (e.g., supply chains, airlines, or traffic flows).

Training large-scale AI models is becoming increasingly difficult and costly for

classical computers. Two promising areas where QCs are expected to help drive AI
even further forward are Natural Language Processing and Artificial Neural
Networks.

QCs could offer a big leap forward for drug discovery and medical services. This is
because drug molecules are quantum mechanical systems themselves, meaning
that QCs are inherently more suitable to simulate them than classical computers.
QCs are expected to enable better Computer-Assisted Drug Discovery (CADD)
tools and thus reduce current high failure rates, R&D costs, and long development
cycles. For example, QCs could help model 3D protein-folding structures in drug
discovery or help predict the interaction of a drug candidate with multiple biological
targets to provide clues into toxicity, pharmacokinetics, and multi-target action. In
addition, QCs are expected to better analyze the increasing amount of data in the
healthcare industry, leading to improved personalization and precision interventions;
earlier, more accurate, and faster diagnoses; and ultimately, more lives saved.

Supporting climate agendas, QCs may help simulate the complicated chemistry of
prototype battery designs and accelerate the R&D process, supporting the transition
to renewable energy. By boosting efficiency and reducing emissions, QCs could
help discover new catalysts for oil refining; Carbon, Capture, Utilization, and
Storage (CCUS); or ammonia manufacturing. In addition, QCs’ potential ability to
simulate the Earth’s climate more dynamically may enable the more accurate
prediction of natural disasters.

In finance, QCs are expected to improve targeting and prediction, portfolio

optimization, risk management, and fraud detection. The use of QCs is expected to
improve the performance of Monte Carlo-based options pricing, portfolio
optimization, and dynamic arbitrage.

It is thought that QCs could help overcome the limitations of existing analytical
models to sift through large amounts of behavioral data, enabling financial
institutions to offer more personalized products and services to customers in real
time. Quantum-optimized loan portfolios focused on collateral could also allow
financial institutions to improve their offerings, possibly lowering interest rates and
freeing up capital.

Cybersecurity risks from QCs need significant attention. Shor’s algorithm theorized
that a large fault-tolerant QC could break the asymmetric (public-key) encryption
used to secure most of the internet in a fraction of the time required by classical
computers. It has been estimated that a QC capable of breaking today’s RSA
encryption would need around 100,000 times more qubits than today’s best
machines, as well as 1/100th of the error rate. Grover’s algorithm would similarly
need a large fault-tolerant QC but would provide a quadratic speed-up in terms of
attacking symmetric encryption standards like the Advanced Encryption Standard
(AES). However, the idea of “Harvest Now, Decrypt Later (HNDL),” whereby bad
actors may harvest the encrypted data now with the intent of utilizing a QC to
decrypt it in the future, means the threat is in the present. This quantum threat is
akin to the Y2K problem at the turn of the millennium (which had an estimated cost
of $200 billion to $850 billion), although the quantum threat is less well-defined, in
that we do not know when it is going to happen, and countries are unlikely to be
fully transparent about their progress on QC.

The cryptography underpinning Bitcoin and cryptocurrencies in general is also at

risk of being broken. Furthermore, with the increasing investment in Metaverse and
Web3 projects that are intended to build a more decentralized web using distributed
ledger technologies (DLTs), breaking the underlying cryptography could enable
nefarious actors to break claim ownership over digital assets on DLTs. To reduce
this threat and improve security, the cryptocurrency community should agree on and
deploy a quantum-resistant signature scheme.

How Nation-States, Corporates, and Market Participants Can Prepare

Countries are uniquely positioned to prepare for quantum computing due to their
ability to facilitate the long-term funding necessary for the sector to grow to scale.
Governments that engage with industry early are more likely to set themselves up
with a strong foundation in developing quantum computing technologies — the UK
National Quantum Technologies Programme is a good example of this. Countries
that do not invest risk falling behind in a future quantum computing arms race,
similar to that of AI. However, government funding is only useful if it yields results —
we used the number of papers published and patents filed in the space as proxies
of how successfully countries have been investing in QCs so far.

There is currently a talent shortage of quantum computing experts, and as

competition to deliver larger-scale QCs heats up, the talent shortage risks becoming
a significant barrier to the growth of the industry. This is in part because the
quantum computing industry is not yet mature enough that most individuals cannot
do without at least a basic understanding of quantum mechanics. Consequently,
progress in the space until now has been predominantly achieved by PhD-level
scientists with advanced knowledge from disciplines including quantum mechanics,
computer science, and electronic engineering. However, to bridge this gap,
professional development courses and internship programs are emerging. As
products mature, more commercial and industrial skills will be needed, such as
engineering and project management. One additional concern for governments is
that the talent pool in the quantum computing field is global in nature, potentially
creating national security concerns.

Supply chains for QCs could also be of strategic importance to nation-states. The
quantum computing supply chain is as global as that of semiconductors but even
more highly specialized and complex, with different types of QCs using completely
different qubit technologies. It is too early to say if there will even be a winning qubit
technology that achieves adoption across the world (in the same way that the silicon
chip became predominant in the manufacturing of classical computers), or to predict
the impact this may have on a still-developing quantum computing supply chain.
Qubits are susceptible to noise, which comes in the form of ambient thermal energy.
Thus, quantum computing technologies, such as superconducting qubits, need to
be supercooled to extremely low temperatures to prevent decoherence. This need
for QCs to be protected from environmental factors is one of the reasons that, for
many, the cloud will be the main method of access.

Similar to AI, QCs have the potential to reshape our world by enabling
breakthroughs in medicine, material science, finance, and other industries, while
also posing equally great risks if used improperly. For example, breakthroughs in
medicine design also bring potential breakthroughs in the design of chemical
weapons. Shortly after the 2019 National Defense Authorization Act (NDAA)
authorized the Department of Defense (DoD) to create a Quantum Information
Science (QIS) research, development, and deployment program, Congress added a
mandate for ethical considerations in the NDAA, requiring the DoD to develop a
plan for the “development of ethical guidelines for the use of quantum information
science technologies.”

Corporates should also prepare for the era of commercial quantum advantage,
through actions including improving quantum computing awareness within the
company, creating impact assessments at the C-suite and board level, examining
collaboration and partnerships, and contextualizing opportunities. Total addressable
market (TAM) forecasts vary both in terms of their starting number (spanning from
$370 million to $1.1 billion in 2022) and their growth rate (25% to 50%), resulting in
significantly diverging forecasts over time. The use of QCs will mostly be a cloud-
based service in the near- and medium-term, accelerating adoption. On the other
hand, barriers include the complexity of integration with existing technology stacks
and the upskilling of employees.

The encryption methods that both protect both stored data and establish secure
networks are at risk. Protective steps are needed today, as adversaries may already
be harvesting data. The National Institute of Standards and Technology (NIST)
released the third-round result for their Post-Quantum Cryptography (PQC)
standard competition in 2022 and is expected to finalize and publish their standards
by 2024. Public key encryption algorithms will need to be replaced in every platform
that uses them. Given the scale of the challenge and the opportunity, we suggest
the time to act is now.

There has been a significant rise in the amount invested by venture capitalists
(VCs) since the claims of quantum supremacy in 2019 that brought quantum
computing to the attention of the public. Eighty percent of VC investment in
quantum computing has occurred since the start of 2020. Despite the market turmoil
in 2022, quantum computing received more VC investment than ever, with $1.8
billion of investment, equating to around a third of all VC investment to date. That
the most active investors in companies have included government agencies is not
surprising given the importance of digital economies.

Purpose of this Report

This report assumes the reader has no prior knowledge of quantum computing and
is targeted at those hoping to get a better understanding of the burgeoning industry.
It is split into three separate chapters, each with their own purpose.

1. Acts as a “Quantum Computing 101” and explains why now is the time to take
notice.

2. Provides an overview of industries that quantum computing is likely to disrupt

(both negatively and positively) and the kinds of use cases it may have in those
industries.

3. Helps nation-states, corporates, and market participants understand the

quantum computing landscape in order to prepare them for computing’s next
leap.

Quantum Computing 101

Introduction
Quantum computing is coming. It will radically change the way we do things. It may
even alter our very understanding of the world around us and beyond. Its power and
potential are difficult to overstate. As such, we need to be prepared.

Each of the last three decades has had its own computing revolution — the
personal desktop revolution in the 1990s, the mobile revolution of the 2000s, and
the cloud computing revolution of the 2010s. While each gradually changed how we
interacted with computers, none have made us fundamentally rethink what it means
to actually “compute.”

Quantum computing may do exactly that.

Our analysis of third-party total addressable market (TAM) forecasts for quantum
computing range between $700 million and $8.6 billion by 2027, with a compound
annual growth rate (CAGR) of up to 50%. By some estimates, the global Quantum
Computing as a Service (QCaaS) market could reach $26 billion by 2030,
representing an approximate 80% CAGR from 2021.3 A deep dive into investment
flows shows us that over the past few years, the quantum computing market has
received significant investment from venture capital (VC) investors, making it one of
the most densely invested technologies when compared to over 100 other areas of
innovation. In 2022, nearly $1.8 billion was invested into quantum computing
companies by VCs in the private markets (Figure 1).

Figure 1. Venture Capital Invested From 2010 through 1Q 2023 Figure 2. Range of Current TAM Forecasts (2022-27E)
90 ($bn)
Angel and Seed Early-stage VC Later-stage VC
80 10.0

70 9.0
8.0
60
7.0
50
6.0
40 5.0

30 4.0
3.0
20
2.0
10
1.0
0.0
2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 1Q 2022 2023 2024 2025 2026 2027
'23

Source: PitchBook Data Inc., Citi GPS Source: Citi GPS, Multiple TAM Forecasts

Inflection Point

Computing is at an inflection point. After decades of consistent increases, the

growth rate of computing power is reaching its physical limit. At the same time,
quantum computers are edging closer to generating commercial value. This
presents both a potential opportunity for early adopters and a potential risk for
others. In a number of industries, even a small advantage in computational power
could be business critical.

3 The Quantum Insider, “Quantum Computing as a Service Market to Hit $26 Billion by

End of Decade,” August 12, 2021.

Why Quantum Computing May Be the Defining

Technology of the 2020s
Since access to quantum computing on the cloud was launched in 2016 by IBM, a
number of competitors have begun to offer Quantum Computing as a Service
(QCaaS). The demand for such services is increasing rapidly, with 74% of large
global enterprises having some form of plan to adopt quantum computing,
according to a survey by Zapata Computing.4 This reflects the fact that 48% of
executives believe quantum computing will play an important role in their
organization as early as 2025.5

Figure 3. Anticipated Quantum vs. Classical Computing Power Over Time

Source: Citi GPS

The advancement in semiconductor circuit fabrication over time — or the doubling

of transistors per chip every 1.5 to 2 years — is dubbed Moore’s Law. According to
some observers, its tenets have been under threat for the past decade and the rate
of progress in the current non-quantum computing (otherwise known as “classical
computing”) paradigm is expected to tail off in the coming years.

4 Zapata Computing, The Second Annual Report on Enterprise Quantum Computing

Adoption, January 11, 2023.
5 EY, How Can You Prepare Now for the Quantum Computing Future?, June 2022.

The capabilities of quantum computers, on the other hand, are advancing and doing
so at an increasing rate. Many believe we have passed the point of quantum
supremacy following Google’s claim of achieving this milestone in 2019, as well as
further such claims since then by researchers in China and at Canadian firm
Xanadu.6 Quantum supremacy, coined by John Preskill in 2012, defines the point at
which we can perform tasks with quantum computers beyond the capabilities of
classical computers.7 Something that is often misunderstood about these claims is
that these are benchmarking tasks rather than commercially useful ones. In other
words, the tasks were specifically chosen for the sole purpose of demonstrating the
quantum computer’s computational superiority. They otherwise do not have any
functional use and are more akin to the quantum computer executing a randomly
chosen sequence of instructions.8

The next major milestone in the development of quantum computing is that of

commercial quantum advantage. This term describes when quantum computers will
be able to offer practical advantages in solving a valuable problem, whether that is
by solving it faster, cheaper, or more efficiently than any available classical
solution.9 One of the biggest technical challenges in solving the more complex real-
world problems to reach this point, is the high levels of noise current quantum
processors have. In 2023 however, IBM used error-mitigation techniques to
compensate for their noise problem and solve a simplified model of a material — an
experiment that IBM claims suggests that quantum computers could have useful
real-world applications within two years. Regardless of any specific timeframe, this
experiment is being considered by many as a key proof-of-principle that quantum
computers could soon provide commercial quantum advantage in the near term.10
Also, for simplicity going forward, the term “quantum advantage” will be used in this
report to refer to commercial quantum advantage.

Figure 4. Generations of Computing

Central Graphics Tensor Quantum

Processing Unit Processing Unit Processing Unit Processing Unit
(CPU) (GPU) (TPU) (QPU)

Source: Citi GPS

6 Elizabeth Gibney, “Hello Quantum World! Google Publishes Landmark Quantum

Supremacy Claim,” Nature, October 23, 2019; Matthew Sparkes, “Quantum Supremacy
Has Been Achieved by a More Complex Quantum Computer,” New Scientist, September
21, 2021; Lars S. Madsen et al., “Quantum Computational Advantage with a
Programmable Photonic Processor,” Nature, Vol. 606, June 2022.
7 Edwin Penault et al., “On ‘Quantum Supremacy’,” IBM, October 21, 2019; John Preskill,

Quantum Computing and The Entanglement Frontier, California Institute of Technology

Institute for Quantum Information and Matter, November 13, 2012.
8 John Preskill, “Why I Called It ‘Quantum Supremacy’,” Quanta Magazine, October 2,

2019.
9
W. J. Zeng, “Clarifying Quantum Supremacy: Better Terms for Milestones in Quantum
Computation,” Medium, January 31, 2019.
10 Davide Castelvecchi, “IBM Quantum Computer Passes Calculation Milestone,” Nature,

June 14, 2023.

With the increasing pace that the capabilities of quantum computers are growing at,
the shift to quantum advantage is likely to be sudden. At the same time, due to the
inherently specialist nature of quantum computers, the step change to quantum
advantage should come at different times for different industries and use cases.
Furthermore, quantum computers are also not likely to be a replacement for our
current classical computing infrastructure, but rather an invaluable addition to it.
After the point of quantum advantage, we’ll have leapt into what many will consider
the next era of computing.

Understanding Quantum Computers

The Basics
A quantum computer is a machine that uses the quantum states of matter such as
atoms or superconducting circuits to perform calculations. They are inherently
probabilistic machines — quantum computers output a sample of a probability
distribution. In contrast, the classical computers we know and use today are
deterministic machines, meaning that they operate using binary computation of only
1s and 0s and provide an exact answer.

The probabilistic nature of quantum computers makes them incredibly powerful

when it comes to solving some types of problems. As they grow in capability, they
are expected to improve our computational abilities and solve problems that would
otherwise be impossible to address on our current classical computers. We look into
this further in our “Areas Quantum Advantage Could Occur” section below.

However, that same probabilistic nature of quantum computers means they are only
good at certain tasks and not likely to replace classical computers in the medium
term. Instead, they will augment the current computational infrastructure.

How They Work

Quantum computers process information by harnessing the principles of quantum

mechanics — the laws that govern the behavior of atoms and photons. By isolating
these particles and finding ways to manipulate them, scientists can create the basic
building block of a quantum computer — typically a two-state quantum bit or “qubit.”
Qubits can simulate their classical counterparts — occupying a state of either 0 or 1
— but can also be in any combination of both states at the same time through the
quantum mechanical property of “superposition.” Qubits are often represented by a
“Bloch sphere,” as shown on the left side of Figure 5.

Figure 5. Diagrammatic Representation of a Quantum Bit (Qubit) vs. a Classical Bit

0 0

1 1
Source: Citi GPS

Superposition is the ability of a qubit to represent both 0 and 1 simultaneously.

Imagine that a qubit is a coin spinning on a tabletop. As it spins it can be considered
a probability function, with a chance of being heads or tails when it stops spinning. It
will continue to be a probability function of possible results until it falls or is stopped,
and the result is measured.

Another quantum mechanical property known as “entanglement” occurs when two

quantum systems interact in such a way that their states cannot be described
independently. This unique quantum behavior binds the destiny of a set of different
particles so that what happens to one will affect the others. Entanglement, which
was once described by Albert Einstein as “spooky action at a distance” (and was the
basis for the 2022 Nobel Prize in Physics), seemingly makes it possible to
manipulate both states of the qubits simultaneously. So, instead of performing a set
of calculations one after another, a quantum computer could perform them all at the
same time.

To perform calculations, a quantum computer must be able to put qubits into a state
of superposition and manipulate the superpositions. The length of time that a
system can maintain this state is called the “coherence” time. Just as the spinning
coin from the above example will eventually stop due to friction, qubits are
susceptible to noise — something that raises its own challenges.

Although spinning coins is a useful analogy for conceptualizing quantum

mechanical concepts, the real science is far more complex. For instance, there are
numerous models for building a quantum computer, something we also provide a
high-level description of in the “The State of the Quantum Computing Market”
section below.

A Highly Simplified Example

One very simplistic way of attempting to conceptualize how a quantum computer

processes information is to use the example of probability distributions being added
together. The deterministic nature of a classical computer means that when asking it
to add two numbers, say 9 + 11, it will output the exact same answer each time —
20. In contrast, a quantum computer works with probability distributions, so it would
input two probability distributions centered at 9 and 11 and output a probability
distribution centered at 20, meaning that while the quantum computer would tell you
9 + 11 = 20 the vast majority of the time, it may also tell you it’s equal to 18 or 22 on
other occasions. To obtain the correct result with more certainty, the quantum
algorithms can be run multiple times, generating a probability distribution.

Figure 6. Probabilistic vs. Deterministic Computing

Source: Citi GPS

That said, probability distributions added together is an incredibly simplified

illustration of how quantum computers operate. Quantum computers are orders of
magnitude more complex than this, and a complete explanation requires an
understanding of trigonometry, differential equations, and numerous areas of
mathematics beyond the scope of this report.

Nobel laureate and famed physicist Richard Feynman is often quoted as saying that
“If you think you understand quantum mechanics, you don’t understand quantum
mechanics.”11 And while there is some truth to this statement, in that quantum
mechanics is probably as divorced from our lived reality as a field of science can be,
it risks potentially painting quantum physics (and hence, quantum computing) as a
mystical, esoteric, and incomprehensible subject. If one is able to put one’s
knowledge of the classical world and lived reality aside and evaluate quantum
computation purely using mathematics, the topic is considerably more
approachable.

11 New Scientist, “Quantum Mechanics,” accessed March 6, 2023.

The Rapid Development of Quantum

Computers
Many people ask why there needs to be a discussion about quantum computing
when the point of quantum advantage has yet to be reached. The answer is
because the era of quantum advantage is likely to arrive relatively suddenly. Why?
In short, because quantum computers have exponential scaling power.

This, alongside the fact that quantum advantage, by definition, comes after quantum
supremacy, means that by the time that quantum computers can offer a practical
advantage to businesses in solving valuable problems, they will be widening the
gap to their classical counterparts at pace.

One of the consequences of quantum mechanics is that by just adding one

additional qubit to a quantum computer, you can double the total number of
potential states the quantum computer can be in at the same time. Given that the
number of states a quantum computer can occupy at any one time is often
considered proportional to the computational capabilities of that machine, we can
see how quantum computing may very easily overtake classical computing. As
Figure 7 and Figure 8 show, the exponential nature of quantum computers means
that, as they grow in size, they can very quickly have more potential states that can
simultaneously exist than there are atoms in the observable universe!12

Figure 7. Exponential Scaling of Qubits Figure 8. Visualization of Exponential Scaling of Qubits

1200
Number of Number of Potential States That Can
Number of Potential States That

Qubits (n) Simultaneously Exist (2n) 1000

Can Simultaneously Exist

n=1 (21) = 2
n=2 (22) = 4 800
n=3 (23) = 8
600
n=4 (24) = 16
n=5 (25) = 32 400
… …
(2100) = 1,267,650,600,228, 200
n=100
229,401,496,703,205,376
0
1 2 3 4 5 6 7 8 9 10

Source: Citi GPS Source: Citi GPS

In other words, to match the rate of progress seen in classical computing for some
applications (i.e., the doubling of classical computing power observed in Moore’s
Law), the number of qubits in a quantum computer would only, in theory, need to
increase by one every two years. However, we are seeing considerably faster
progress.

Google’s observations suggest that the rate of progress in the industry could be
described as doubly exponential growth. As Scientific American reported, in
December 2018, Google was able to replicate calculations run on their best
quantum computer using a normal classical laptop.13

12 Bysome estimates, the number of atoms in the observable universe is around 1078 to
1082
13 Kevin Hartnett, “A New ‘Law’ Suggests Quantum Supremacy Could Happen This

Year,” Quanta Magazine, June 21, 2019.

By January 2019, however, they had managed to improve their quantum computer
to the point that a powerful classical desktop computer was needed to replicate the
results. This rate of progress continued to the point that by February 2019, to
replicate the results of their then-most advanced quantum computer, Google had to
instead use their incredibly powerful server network.

This rate of doubly exponential improvement is referred to as “Neven’s Law” after

Hartmut Neven, Director of the Quantum Artificial Intelligence Lab at Google.
Neven’s Law, like Moore’s Law, is an empirical law — one that has come about from
observation. Hence, whether Neven’s Law will hold true in the long term remains
unclear. It depends a lot on the progress of the industry and whether it is
fundamentally possible to keep control of qubits as we build larger quantum
computers. It can be neither proven nor ruled out.

Figure 9. Neven's Law of Doubly Exponential Growth vs. Moore's Law of Exponential Growth

Computing Power (log-scale)

Time

Exponential Growth Doubly Exponential Growth

Source: Citi GPS

What this means is that given the rate at which quantum computing technologies
are maturing over time, the increase in computing power is exponentially increasing
even when plotted on a logarithmic graph (Figure 9). Neven explained that with
quantum computers’ doubly exponential growth, “it looks like nothing is happening,
nothing is happening, and then whoops, suddenly you’re in a different world.”

We can see from the figure above how quantum computers following such a path of
progress could easily catch up to and overtake classical computers, despite the
latter’s multi-decade head start.

Areas Quantum Advantage Could Occur

There are four broad areas where quantum computers can provide an advantage.
We dive deeper into how these for areas will impact different industries and discuss
practical use cases later in the report.

Figure 10. Areas of Anticipated Quantum Advantage

Source: Citi GPS

− In optimization, quantum computers are likely to outperform their classical

counterparts when solving such problems. This will lend itself to any use case
where there is an efficient answer to a problem with a high number of potential
solutions, including portfolio optimization and route planning.

− For machine learning, quantum computers may accelerate existing workloads

or unlock entirely novel methods outright. This is likely to have an impact on the
likes of fraud detection and the development of artificial intelligence.

− Regarding the simulation of molecules and other quantum systems, quantum

computers are expected to have an exponential speed-up over classical
computers. As legendary physicist Richard Feynman put it, “Nature isn't
classical…if you want to make a simulation of nature, you'd better make it
quantum mechanical.”14 Quantum simulation is one of the areas of greatest
potential for this technology due to the number of areas that rely on the
simulation of chemical reactions.15 This includes the likes of fertilizer production,
battery technology, drug development, and the manufacturing of advanced
materials.

− In cryptography, quantum computers of sufficient size and operating efficiency

pose a threat to existing cryptographic standards, which is in fact what really
launched the widespread interest in quantum information and computing.
Fortunately, there are efforts by the National Institute of Standards and
Technology (NIST) and the cryptographic community to introduce new quantum-
resistant protocols. The challenge will lie in ensuring their broad adoption across
the globe.

14 Andreas Trabesinger, “Quantum Simulation,” Nature Physics, Vol. 8, No. 263, April

2012.
15 John Preskill, “Quantum Computing in the NISQ Era and Beyond,” Institute for

Quantum Information and Matter and Walter Burke Institute for Theoretical Physics,
California Institute of Technology, July 30, 2018.

While the above-mentioned areas will affect a broad range of industries, it is

important to note that quantum computers, given their highly specialized nature, will
only impact discrete areas within each industry at first. In other words, quantum
advantage will reveal itself over time far more granularly than at an overall industry
level. It will probably first occur in very specific use cases, such as modeling a
particular type of protein or running a particular type of machine learning algorithm
better than a classical computer. In part due to the granular nature of quantum
advantage, a hybrid approach to computing looks inevitable.

Regarding which of the four areas of quantum advantage will appear first, this
question is still open to debate — most companies and experts we spoke to tended
to have relatively strong convictions in their chosen area of quantum advantage.
What we can say is that due to the need for error correction and the large number of
qubits required to break today’s cryptography standards, quantum advantage in
cryptography looks likely to be the last.

Regardless of where commercial quantum advantage is first observed, when it

occurs, quantum computing may very well be the future of computing in that area.

We spoke to Professor David J. Wineland, who won the 2012 Nobel Prize in
Physics for his “ground-breaking experimental methods that enable measuring and
manipulation of individual quantum systems” and is currently a Philip H. Knight
Distinguished Research Chair in Physics at the University of Oregon, to get his
insights on the future of quantum computing.

Expert Interview with Nobel Laureate, Professor David J.

Wineland, at the University of Oregon
Q: You won the Nobel Prize for your work in manipulating quantum systems
— something that many consider a milestone in the development of quantum
computing. What initially attracted you to the field and to what extent was
quantum computing on your mind?

Prof. Wineland: Like many others, my interest in quantum computing really started
back in 1995 after Peter Shor published his now-famous factoring algorithm. At first,
I think that got a lot of people interested in the idea of a quantum computer because
of its potential impact on cybersecurity. Not surprisingly, government agencies also
became very interested in this and there was a sudden influx of support for
investigating quantum computing. As you point out, however, my 2012 Nobel Prize
wasn’t technically for quantum computing, but rather for developing experimental
methods that enabled the measuring and manipulation of individual quantum
systems (although, of course, this is inherently important to quantum computing). As
such, I can’t say building some kind of large universal quantum computer was
Professor David J. Wineland something I’d thought of day-to-day during my research. Rather, back then, and as
Nobel Laureate and Philip H. Knight is still the case, my work and that of the group I was involved with, was focused on
Distinguished Research Chair in Physics, demonstrating the simple primitives of quantum computing as well as quantum
University of Oregon “gedanken,” or thought, experiments that were posed by Einstein and his
colleagues in the early days of quantum mechanics. So rather than aiming to build a
quantum computer, we hoped to make useful contributions to the general problem
in the long run, such as through the development of better quantum gates.

What I find particularly interesting about the study of quantum systems is that, even
now 100 years on, we are still addressing questions that the founding fathers of
quantum physics asked themselves, such as the postulated Einstein–Podolsky–
Rosen (EPR) type experiments. An example of an EPR experiment is that when you
entangle two atoms and then separate them by a large distance, their entangled
state implies that measuring the first atom results in an effectively instantaneous
impact on the second atom such that its measured state is correlated in a precise
way to the measured state of the first atom. And the amazing part is that the
measured state of the first atom is completely random, but the correlations between
the measured states of the two atoms are preserved and occur instantaneously.
These instantaneous correlations have now been unambiguously demonstrated
with the quantum states of photons. At first glance, it appears as if these two
quantum entangled atoms are communicating faster than the speed of light and
thus violate a law of physics — that information cannot travel faster than the speed
of light. But because the measurement outcome of the first atom is random, no
information can be transferred. However, we do know that these correlations are
being transferred faster than the speed of light and I just find that fascinating, and I
don’t think anyone has a satisfactory answer as to why this occurs.

In fact, Einstein himself is often quoted as describing this as “spooky at a distance,”

and it very much is. I think the fact that we’re still trying to find an answer as to why
nature works in this way a century later is very interesting to me.

Q: What do you think of the prospects of building a “universal” quantum

computer?

Prof. Wineland: With respect to the idea of building a large-scale fault-tolerant

quantum computer (often referred to as a “universal” quantum computer), especially
one which would have the capacity to factor cryptographically significant numbers
that underpin much of cybersecurity, I think we’re a long way off yet. We are still in
what is sometimes referred to as the Noisy Intermediate Scale Quantum (NISQ)
computing era, where quantum computers are characterized by noisy gates and
lack the error-correction needed to run complicated algorithms. To move into the era
of Fault-Tolerant Quantum Computing (FTQC), in which we can truly consider
having a universal quantum computer, we will need machines with significantly
more qubits than even the most advanced quantum computers have today, and the
necessary error-correction needed for these complicated quantum algorithms.

That’s not to say it won’t happen though. Even as far back as the mid-90s, I
wouldn’t say there was anyone suggesting building such a universal quantum
computer one day was impossible — it was just that our technology back then was
not good enough for the precise manipulation of atoms or superconducting qubits —
and in fact, we still haven’t mastered control of these. The current challenges
involved in creating a universal quantum computer are still difficult, but I’m optimistic
in the long run. As to when this may happen, however, I couldn’t really say.
Quantum computing has been a technology that’s often been oversold, so I would
caution against over-optimistic forecasts. Back in 1995, when I was at the National
Institute of Standards and Technology (NIST) and we’d learned of Peter Shor’s
algorithm, many of us thought that we’d have made much more progress by now
almost 30 years later. However, I do find it difficult to believe we won’t eventually be
able to create a universal quantum computer.

Q: As well as the risk to cryptography, we identified three other areas of

potential “quantum advantage”: optimization, machine learning, and
molecular simulation. What are your thoughts on quantum computing being
practically useful in these areas in the nearer term?

Prof. Wineland: There are definitely a lot of scientists out there looking at various
uses for quantum computers beyond just creating a large factoring machine. One
particular area of quantum advantage that could have a significant impact on
society is that of molecular simulation.

All molecules at the smallest of scales are inherently quantum mechanical, and
these characteristics are something that quantum computers are anticipated to be
far better at modeling than our current classical computers. As Richard Feynman is
often referred to as putting it: “in order to understand quantum systems, you need a
quantum computer.” In the nearer term, it is very plausible that if quantum
computers continue to develop at their current rate, one of the areas they could
potentially be of great value would be the simulation of molecules that may be
useful in drug therapy. Rather than having to synthesize a new drug in a lab,
scientists may be able to simulate the action of a new molecule on a quantum
computer to verify or nullify its usefulness.

A lot of people feel that, well before quantum computers are able to factorize
cryptographically significant numbers (i.e., the extremely large numbers that are
used in internet encryption), the first real application of quantum computers will be
in one or more of the areas mentioned or for applications that haven’t yet been
considered. I also think there are a lot of areas in quantum information theory that
are very interesting, and some that are already aiding classical computing.

A Hybrid Approach to Computing Is Inevitable

After the point of quantum advantage, we anticipate a combination of classical and
quantum computing is highly likely for the foreseeable future.

Quantum Complexity Theory

Quantum complexity theory is a sub-field of computational complexity theory that

looks to determine the “hardness,” or difficulty, of certain computational problems.
Hardness is defined as the allocation of resources needed by computational models
to solve certain problems. In other words, it is asking how the difficulty of a problem
increases as the size of the problem set increases. Unsurprisingly, the ways
different types of problems overlap are quite complex. Figure 11 illustrates this in a
simplified graphical form:

Figure 11. Example of How Problems Can Be Categorized by Their Quantum Complexity

Source: IBM Quantum, IBM Corporation

Quantum computers and their classical counterparts excel at different types of

problems. Based on current assessments, there will always be areas of computing
where quantum computers do not provide a material speed advantage over
classical computers. Thus, as illustrated in Figure 11 above, the vast majority of
applications of quantum computing that offer a competitive advantage in a given
industry, are likely to occur in the dark blue area. If only part of an overall problem is
solved faster by a quantum computer, then it makes sense to only allocate that
specific piece to a quantum computer.

The Practicalities: Error-Correction and Hybrid Algorithms

Another reason to believe we will always need classical computers is more practical.
The error-correction that quantum computers will need to grow in scale requires
classical computer co-processors. This means that even if in a quantum-enabled future,
we reach a point at which quantum computers are far superior at calculations to
classical computers (and perhaps are even one of the main sources of computing in the
cloud), quantum computers will always need classical computers surrounding them in
order to operate. Figure 12 is an illustration of how such a setup may work.

Figure 12. How Quantum Computers and Classical Computers Will Work in Parallel

Source: Citi GPS

Many algorithms intended to run on quantum computers are, in fact, hybrid

quantum-classical algorithms that will always require a classical computer by
design. Algorithms such as the Variational Quantum Eigensolver (VQE) and the
Quantum Approximate Optimization Algorithm (QAOA) use the quantum computer
to explore the space of possible solutions and classical machine learning to
optimize search parameters.

A potential way of thinking about quantum computing’s future path is to compare a

quantum processor to a classical graphics processor. Analogous to how a Graphical
Processing Unit (GPU) is often a key addition to a high-end personal computer,
providing a significant speed-up on the specialist tasks that the standard Central
Processing Unit (CPU) cannot handle as efficiently, a Quantum Processing Unit
(QPU) may likely be a key addition to the commercial High Performance Computing
(HPC) sector.

We spoke to Professor Sougato Bose, a leading physicist and researcher in

quantum computation at University College London (UCL), whose work featured in
New Scientist magazine, to get his insights on how technology may be used.16

16Thomas Lewton, “The Quantum Experiment That Could Prove Reality Doesn't Exist,”
New Scientist, November 3, 2021.

Expert Interview with Physicist, Professor Sougato Bose,

at University College London
Q: How likely is a quantum-classical hybrid approach to computing?

Prof. Bose: It is almost certain that a hybrid approach to computing will have to be
pursued. Classical computers and processors are robust and operate under
minimally demanding circumstances (you use your laptop and mobile phone
anywhere). Quantum computers require exceptional protection from the
environment and will require dedicated centers. Thus, we will access these only for
special problems or as subroutines of various problems, with the rest of the
computation taking place on classical computers. This is why there are many
current algorithms that are hybrid by construction. Additionally, to control and error-
correct quantum computers, we require classical computers.

Q: What impact do you think quantum computers could have in our

understanding of the universe?

Prof. Bose: It is our current understanding that the universe is made from
Professor Sougato Bose
interacting quantum components — so it is a giant quantum computer in a certain
Physicist and Researcher in Quantum
sense, albeit not the fully controllable and programmable one we are building in the
Computation, University College London
laboratories for usage. So far, humans have tried to comprehend the universe using
simplified assumptions and approximate solutions, and the great deal that has been
achieved has been based on the ingenuity of our approximations. Simulations allow
us to go beyond these solvable pen-and-paper limits, but classical computers are
not ideally suited to this task as the components of the universe are quantum, and
typically, you are required to apply exponentially large matrices to exponentially
large vectors.

Thus, quantum computers will enable us to probe the "complexity" frontier in our
understanding of the universe — for example, phases of matter with complicated
patterns of quantum entanglement. This may, in turn, help us in understanding
phenomena such as high temperature superconductivity or even the emergence of
space and time. On a different front, quantum computers, or at least elements
developed for quantum computing, could be incorporated in sensor devices, and
help with higher-precision sensing of various fundamental particles and forces,
which will, again, deepen our understanding.

Q: What do you think that about the current level of quantum education in
terms of the potential workforce in the industry?

Prof. Bose: I think the current level is below what will soon be required by society.
At universities, quantum mechanics itself is typically taught, at least in the form
needed for quantum computation, to undergraduates taking physics as a major, only
in the last or third year. It has not spread out to the curriculum of other disciplines,
as well as to earlier levels of physics curriculum in terms of the very fundamentals
such as qubits and gates. I think this will be a change that will be necessary to have
a larger volume of educated workforce in this area — so, we should introduce
quantum mechanics in the form of quantum computation earlier, as well as use
quantum computation as the example to learn quantum mechanics.

The State of the Quantum Computing Market

Quantum computers are often mistakenly thought of as being just a homogenous
architecture, in the same way that all classical computers rely on silicon transistors.
However, in quantum computing, two different fundamental architectures are being
explored.

Types of Quantum Computers: Gate-Based vs. Quantum Annealing

The most widely explored model for building a quantum computer is that of a gate-
based quantum computer. As the name suggests, this model involves the usage of
a set of universal gate operators to execute circuits that outline the desired
calculation. This model is applicable to numerous general-purpose use cases, as it
can run various types of algorithms efficiently. This is one of the reasons that most
of the hardware development in the quantum computing industry is currently being
done on gate-based quantum computers. As such, most of the algorithmic research
being done today focuses on being applied to this type of machine.

Figure 13. Example of Quantum Logic Gates

Source: Wikimedia Commons, licensed under the Creative Commons Attribution-Share Alike 4.0 International
license

A second model — a quantum annealer — is a device specialized in solving

optimization problems. It works by harnessing a process called annealing to find the
lowest energy state of a system corresponding to the most optimal solution to a
specific problem. For this reason, it is particularly good at solving quadratic
unconstrained binary optimization (QUBO) problems.

Figure 14. Visualization of the Quantum Annealing Process

Source: D-Wave Systems Inc.

Different Qubit Technologies

Today, numerous methods of building physical qubits for gate-based quantum

computers are being explored. This is in part because the quantum physics a
quantum computer attempts to exploit can be seen in various mediums of matter
and energy. So, different teams around the world have chosen different physical
starting points to build a quantum system into a quantum computer.

Examples of qubit technologies include:

 Trapped Ions

 Superconducting

 Photonic

 Neutral Atoms

 Quantum Dots

 Nitrogen Vacancy Centers

Where We Are Today

Decades of R&D and billions of dollars have gotten the quantum computing industry
to where it is today, and it is only now we are beginning to see the fruits of that
labor.

Physical quantum computers (QCs) exist today and have anywhere up to 433
physical qubits. Despite the qubits being physically very small, the overall amount of
space QCs need is typically anywhere from around the size of a cupboard to a
small room, due to all the apparatuses needed to control the physical qubits. Far
more important is the actual number of physical qubits and high error rates of these
early machines — something that led quantum physicist John Preskill to dub this
era of hardware Noisy Intermediate-Scale Quantum (NISQ).

As John Preskill put it, when he described the NISQ era, he was “imagining
quantum computers with noisy gates unprotected by quantum error-correction.”17
The term “noisy” refers to the imperfect control scientists currently have over qubits,
resulting in the noise introduced into quantum systems that results in QCs being
error-prone today. The term “intermediate scale” references the size of current
quantum computers as measured by their qubit count, expected to range from 50 to
a few hundred physical qubits in the near term.

Currently, NISQ-era QCs are measured in terms of the number of physical qubits,
as opposed to the number of logical qubits. Logical qubits can be described as
multiple physical qubits that are operated in such a way that they are guaranteed to
retain their information without errors. The most prominent technique is to create a
circuit of physical qubits that output a single error-corrected result. However, this
technique and other error-correction techniques only work if certain qubit error rates
are low enough to begin with. The proposed ratio of physical to logical qubits varies,
but with current designs, it can be anywhere from around the order of 1,000:1 to
10,000:1.

17John Preskill, “Quantum Computing in the NISQ Era and Beyond,” Institute for
Quantum Information and Matter and Walter Burke Institute for Theoretical Physics,
California Institute of Technology, PDF, July 30, 2018.

Notably, the qubit count is by itself not necessarily the best way to compare
quantum computers; the quality of those qubits is just as important. Simply put, a
QC with 50 high quality (i.e., low-noise) qubits may very well perform calculations
far better than a QC with 5,000 low quality (i.e., high-noise) qubits. In addition to
differences in qubit quality, the variation in underlying qubit technologies makes
direct comparisons between different QCs very difficult. Factors such as clock
speed, operational speed, and connectivity between qubits also have a significant
impact. To truly determine how advanced and useful a QC is, additional metrics
beyond the scope of this report, such as the error rates and the connectivity of
qubits, need to be considered. However, even all these metrics do not tell the full
story.

IBM’s recent proof-in-principle of quantum advantage — where they solved a

simplified (unrealistic) model of a material — was in fact undertaken on their older
127-qubit quantum processor from 2021. For this experiment, they described
running calculations involving all 127 qubits and “up to 60 processing steps — more
than any other reported quantum-computing experiment.”18 They then used an
error-mitigation technique where they measured the noise in each of their qubits
and extrapolated back to what these measurements would have looked like in the
absence of noise. Ultimately, in doing so, IBM showed just how important employing
the right techniques and algorithms are, regardless of the qubit count of the QC.

However, to provide some context on the state of the quantum computing market,
the table in Figure 15 highlights the qubit counts of QCs from different companies
based on publicly available information. As it stands, IBM is currently leading the
qubit race with their 433-qubit QC announced in 2022. Pasqal then follows with their
324-qubit QC.

Figure 15. Qubit Counts for Some Quantum Computing Companies (as of June 2023)
Qubit Technology Company No. of Qubits
Superconducting IBM 433
Cold Atom Pasqal 324
Cold Atom QuEra 256
Photonic Xanadu 216
Superconducting Rigetti 84
Superconducting Google 72
Trapped Ion IonQ 32
Trapped Ion Quantinuum 32
Source: Company Reports, Citi GPS

While it is impossible to precisely predict the roadmap to our quantum future, we

have investigated the plans of some of the top players in the sector. For instance,
Google intends to have a 1 million physical qubit QC before the end of the decade.
However, to scale the qubit count of individual QCs to this level, it is expected by
some to result in QCs occupying increasing amounts of floor space.19 While this
potentially presents its own challenges, it is not unlike the early era of classical
computers, which have advanced and shrunk in size immeasurably in the decades
that followed. This challenge is unlikely to hinder the use of QCs, as we expect the
vast majority of end-users are likely to use the same technology used to access
their classical computers today — the cloud. We discuss this further in the
“Collaboration and the Cloud” section.

18 Davide Castelvecchi, “IBM Quantum Computer Passes Calculation Milestone,” Nature,

June 14, 2023.
19 Simon Benjamin, “Separating Quantum Hype From Quantum Reality,” Financial

Times, September 2, 2022

Figure 16 shows, in no particular order, the current publicly-available qubit

roadmaps of various companies — something regularly subject to change. It is also
important to note that many companies choose not to disclose their qubit roadmaps
for strategic reasons. The below table was put together from publicly available
information to, again, provide some context to the market. Regardless of these qubit
roadmaps, we found that most companies feel they will be able to provide quantum
advantage to businesses sometime this decade.

Figure 16. Qubit Roadmaps for Some Quantum Computing Companies (as of August 2022)
Company Qubit Tech 2023 2024 2025 2026 2027 2028 2029 2030
ColdQuanta Cold Atom 1,000+
Google Superconducting 1mn+
IBM Superconducting 1,121 1,386+ 4,158+ 10K-100K
Pasqal Cold Atom 1,000+
Rigetti Superconducting 84 336 1,000+ 4,000+
Source: Company Reports, The Quantum Insider, The Verge, Citi GPS

The list above is also by no means exhaustive. For instance, one estimate by the
Quantum Computing Report identified over 106 organizations working on 135
projects in the quantum computing space.20 In addition, in conversations with
corporates, nearly all companies developing a QC have opted to solely pursue a
gate-based structural model, with the exception of companies like D-Wave, which
are pursuing the development of both gate-based and quantum annealing QCs. As
quantum annealers are very different from gate-based QCs, they are not included in
the tables above.

The Challenges We Face

One pressing question in the field is whether NISQ machines are capable of
achieving quantum advantage (which we describe earlier as the point at which QCs
can offer practical advantages in solving a valuable problem, whether that is by
enabling faster, cheaper, or more efficient solutions than classical computers).
There is notable debate surrounding this topic, but most companies we spoke to
feel NISQ machines have the potential to achieve quantum advantage in one or
more of the areas identified earlier. One of the goals of the NISQ era is to either
prove or disprove this conclusively and this effort will require the right combination
of hardware development and algorithm design, as well as an application to the
right problem.21

In terms of hardware design, there are still many challenges that lay ahead for
developing better QCs in the NISQ era. The primary challenge is qubit quality —
manufacturers can create many qubits; however, controlling them all at once is a
problem. Another key challenge is the size of QCs (i.e., the number of qubits that
make them up) — current QCs do not have enough qubits of a high enough quality.

High levels of noise pose yet another key challenge for QCs. For QCs to perform
calculations, they must be able to maintain their qubits in a superposition of states
for a reasonable amount of time, known as the “coherence” time. Similar to how a
spinning coin is subject to friction, qubits are susceptible to noise, which can come
in the form of ambient thermal energy or electromagnetic interference, for example.
This ultimately leads to “decoherence,” which is when the calculations become
unreliable as the noise introduces errors and faults into the system.

20 Quantum Computing Report, “Homepage,” accessed March 6, 2023.

21 KishorBharti et al., “Noisy Intermediate-Scale Quantum (NISQ) Algorithms,” Reviews
of Modern Physics, Vol. 94, No. 1, February 2022

The process of error-correction attempts to mitigate or outright eliminate noise-

induced errors in a quantum circuit through the introduction of additional error-
correcting qubits to the system. Mitigating the impact of noise in quantum systems
and extending coherence times are some of the leading challenges in working
towards entering a future era of Fault-Tolerant Quantum Computing (FTQC)
machines.

The fault-tolerant era is generally considered to be one in which QCs have the
necessary quantum error-correction built in to prevent the quantum errors (or
“faults”) inherent to qubits from cascading through quantum circuits.22 Like classical
error-correction, quantum error-correction involves the allocation of redundant
qubits to undertake the role of correcting errors in other qubits — whether this is
due to the imperfect control we have over them or to environmental interactions.
The idea behind using error-correction is that there will be a point at which the
fundamental accuracy of individual qubits is high enough that the error-correction is
correcting more errors than are being created.23

While we are potentially years away from FTQC machines, most agree that
reaching this stage is required for QCs to reach their full potential and unlock all the
areas of quantum advantage. This is because many quantum algorithms today,
including those that pose a threat to cryptography, require the implementation of
error-correction.

We spoke to Dr. Dario Gil, who is the Senior Vice President and Director of
Research at IBM Quantum, one of the most advanced players in the quantum
computing market, about his opinion on the state of the industry today.

22 Michael A. Nielsen and Isaac L. Chuang, Quantum Computation and Quantum

Information (Cambridge: Cambridge University Press, 2010).
23 Alexandru Paler and Simon J. Devitt, “An Introduction to Fault-Tolerant Quantum

Computing,” downloaded from arXiv, PDF, August 15, 2015.

Expert Interview with Dr. Dario Gil, Senior Vice President

and Director of Research at IBM Quantum
Q: Where is the quantum industry today?

Dr. Gil: Quantum computing is no longer a futuristic concept. We believe we are

entering the Quantum Decade — an era when enterprises will begin to see
quantum computing’s business value. The unprecedented advances in quantum
hardware, software development and services validate the technology’s momentum,
creating an ecosystem that paves the way and prepares the market for the adoption
of this revolutionary technology.

The groundwork for this adoption is being laid by a global ecosystem of more than
400,000 users tapping into the cloud to experiment on IBM’s more than 25 quantum
systems. These users include individuals exploring the technology, to the
developers, scientists, engineers, and domain experts at the more than 200 Fortune
500s, government labs, academic institutions, and startups that make up the IBM
Quantum Network.
Dr. Dario Gil
Our quantum roadmap, first made public in 2020, outlines how the entire stack of
Senior Vice President and Director of
quantum technology, from processor advancement to software modules and
Research, IBM Quantum
services, will support the industry’s progress. Last year, we debuted the 127-qubit
IBM Quantum Eagle chip, our first with more than 100 qubits, and previewed IBM
Quantum System Two. To simulate Eagle, you would need a classical computer
with more classical bits than the number of atoms contained in all of humanity’s 7.9
billion people on Earth. System Two, with its capacity to scale beyond 1,000-qubit
processors, will move us closer to a true quantum data center.

In November 2022, we released the 433-qubit IBM Quantum Osprey, the largest
quantum processor to date, three times larger than the Eagle processor. Like Eagle,
Osprey includes multi-level wiring to provide flexibility for signal routing and device
layout, integrated filtering to reduce noise and improve stability, plus our new high-
density control signal delivery with flex wiring to provide a 70% increase in wire
density and a 5x reduction in price-per-line. We debuted our third-generation control
system, controlling 400 qubits in a single rack at an even lower price point than the
previous generations. And we made strides in our other performance metrics —
quality and speed — with a 4x improvement of quantum volume from 128 to 512
and a 10x improvement in Circuit Layer Operations Per Second (CLOPS) from
1,400 to 15,000 — beating our goal of 10,000 CLOPS.

We also showcased the IBM Quantum System Two, a newly designed “quantum-
centric supercomputer,” that is going to be the path for how we go from the noisy,
small-scale quantum devices of today to the thousands and eventually million-plus
qubit systems of the future. See here for more on IBM Quantum System Two.

Working with our partners, working groups, opensource and researcher programs,
we’ve seen over 1,700 papers published using IBM Quantum and Qiskit technology
in areas including chemistry, finance, and machine learning. For example, our
partners in finance are exploring how quantum computers could help solve risk
analysis, and options pricing problems. Quantum could also help cut through the
complexity of today’s trading environments. Using combinatorial optimization,
quantum could help investment managers improve portfolio diversification, and
rebalance portfolio investments to more-precisely respond to market conditions and
investor goals.

Q: What are the remaining hurdles to achieving quantum advantage?

Dr. Gil: Quantum computers are not a replacement for classical ones. But we know
that certain problems exist that classical computers will never solve. This is why we
need to reach Quantum Advantage: when quantum computers are either cheaper,
faster, or more accurate than classical computers at the same practical task. To
deliver on the promise of Quantum Advantage, we’re committed to improving three
critical quantum system attributes: scale, quality, and speed.

We are rapidly scaling the number of qubits in our systems, as evidenced by the
launch of Osprey, and our plans to deploy a processor with more than 1,000 qubits
in 2023, when we believe it will be possible to explore applications with a Quantum
Advantage. Beyond 2023, we are preparing modular, interconnected systems —
quantum datacenters — for processors with millions of qubits as exhibited by our
design for IBM Quantum System Two.

The quality, or how accurately a quantum computer’s qubits process information, is

also doubling every year. This is obviously a critical element to the interpretation of
the results from quantum computing experiments. But quantum computing must
also be able to complete this useful, practical work in a reasonable amount of time.

We are committed to improving these three attributes, in parallel, with the goal of
finding practical quantum computing use cases.

Q: Where will quantum computing be by the end of the decade?

Dr. Gil: Innovation alone can’t unlock the full potential of quantum computing. At
IBM, we see the continued, rapid growth of a global ecosystem — from individual
students, scientists, developers, and engineers, to organizations and institutions —
being the key to making the leap into the second half of the decade that explores
and develops applications in new materials, drug discovery, supply chain and
macroeconomic optimization, and more. And it will all be done over the cloud on
multiple, interconnected systems with millions of qubits, and frictionless, open
developer tools.

The end of the Quantum Decade will look nothing like the beginning. We’ll be
working with quantum processors with thousands of qubits; we’ll have a whole
workforce with years of experience in quantum, and enterprises will have seen the
payoff of quantum. Any technology leader who isn’t actively building quantum into
their plans risks being left behind.

Industry Impact
As we discussed previously, quantum computing is expected to surpass the
capabilities of its classical counterpart in four main areas: optimization, machine
learning, simulation, and cryptography. With these anticipated areas of quantum
advantage, quantum computers are likely to bring disruptive changes to many
industries.

A survey conducted by the National Quantum Computing Centre (NQCC) and EY in

the UK investigated the views of senior corporate executives on the ability of
quantum computing to disrupt their sectors. The NQCC found that, in the UK,
almost all (97%) of the 501 executives surveyed expected quantum computing to
disrupt their sectors to a moderate or high extent. Furthermore, nearly half (48%) of
the respondents reported thinking quantum computing will play an important role in
their organizations by as early as 2025. To address this, most respondents said that
their firms will be taking concrete steps within the next one to two years to
prepare.24

It is important to be aware of the practical opportunities and risks quantum

computing poses to each industry. As such, in this chapter, we look at some
quantum computing use cases in a few select industries including:

 Manufacturing and Logistics

 Artificial Intelligence

 Healthcare

 Energy and Climate

 Finance

 Cybersecurity

 Cryptocurrency

24
EY, 81% of UK Business Leaders Expect Industry Disruption From Quantum Computing By
2030, According To EY Study,” June 15, 2022.

Manufacturing and Logistics

The Fourth Industrial Revolution (Industry 4.0) Is Already Here
Manufacturing and logistics are widely considered to be among the main early
beneficiaries of quantum computing. The entire supply chain has already entered a
new era, Industry 4.0, where an intelligent network of machines and processes is
adopted for industrial productions with the help of sensor, computing, and
communication technologies.25 We have seen an increasing adoption of cyber-
physical systems (CPS), the industrial Internet-of-Things (IIoT), cloud computing,
robotics, and artificial intelligence (AI) in smart factories.

Consequently, the datasets these systems are built upon are expanding rapidly in
both size and complexity, bringing about challenges to the processing power of our
current classical computers.26 Quantum computing may offer solutions to these
challenges and potentially open the door to a further transition from Industry 4.0 to
an even more interconnected Industry 5.0.

Material Discovery
New materials are regularly discovered in the pursuit of producing better quality
goods that are lighter, stronger, or cheaper. However, despite all our current
industrial manufacturing processes, nature still produces numerous superior
materials. For instance, the teeth of limpets, a type of aquatic snail, can be 13 times
as strong as ordinary steel and are produced at limpets’ body temperature, rather
than the extremely high temperatures and pressures required by many industrial
production processes.27 Unfortunately, today’s classical computers struggle to
model even moderately sized molecules (with the accuracy needed to know how to
recreate them), let alone large-molecule mineral-protein composites like limpet
teeth.

One day, quantum computing may bridge this gap and enable the comprehensive
modeling of sophisticated molecules, dramatically benefitting material discovery.
Once quantum computing’s anticipated advantages in molecular simulation become
mature enough, manufacturing within various sectors, including automotive, airline,
electronics, and military, could benefit from:28

– Materials with both lighter weight and higher strength.

– Batteries with higher efficiency and more durable storage capacity.29

– Catalysts with higher efficiency in fueling various industrial production

processes.

25 Thomas Philbeck and Nicholas Davis, “The Fourth Industrial Revolution,” Journal of
International Affairs, Vol. 72, No. 1, Fall 2018.
26 David Reinsel, John Gantz, and John Rydning, Data Age 2025: The Digitization of the

World: From Edge to Core, IDC (sponsored by Seagate), November 2018.

27 John H. Lienhard, “Engines of Our Ingenuity No. 2996: Tensile Strength & Limpet

Teeth,” University of Houston, accessed April 5, 2023.

28 Rhys Blakely, “Ministry of Defence to Develop Quantum Computer for Use on

Battlefields,” The Times of London, June 9, 2022; Masoud Mohseni et al.,

“Commercialize Quantum Technologies in Five Years,” Nature, Vol. 543, No. 171-174,
March 2017.
29 Jeannette Garcia, “IBM and Daimler Use Quantum Computer to Develop Next-Gen

Batteries,” IBM, January 8, 2020.

Production Processes
As production processes become more digitalized in the era of Industry 4.0,
quantum computing could in many ways help enable better and more efficient
decision making.

One such area is that of “digital twins,” or virtual simulations of products or system
lifecycles that can be updated with real-time data. Although the use of digital twins
has already proven to be an efficient tool for component testing and production
processes, quantum digital twins could enhance production controls by enabling the
visualization of real-time information on product quality in an intuitive form, while
dramatically lowering simulation costs.30 Quantum digital twins could also offer
significant performance gains in testing simulations, enabling them to find the most
optimal strategies in the shortest time.31

Other areas in which quantum computing shows promising potential to increase

efficiency include:32

– Predicting customer requirements and demands based on complex data-

driven simulations.

– Optimizing production targets in real time.

– Managing Automated Guided Vehicle (AGV) and Autonomous Intelligent

Vehicle (AIV) fleets in factories.

Supply Chains
Global supply chains are sometimes not as agile as we expect them to be. World
events ranging from the Suez Canal blockage to the COVID-19 pandemic have
shown how susceptible our supply chain and logistics systems are to sudden
changes in consumer and business demand, raw material availability, and shipping
and distribution channels.

Most logistics systems are essentially optimization problems. When these

optimization problems are applied to the real world, we find that actual situations
become exponentially more complex for every extra variable involved. For example,
introducing additional vehicles, routes, or drivers in a logistics system.33 Moreover,
as we head further into Industry 4.0, supply chains have become more digitalized
and interconnected.

30 C.K. Lo, C.H. Chen, and Ray Y. Zhong, “A Review of Digital Twin in Product Design

and Development,” Advanced Engineering Informatics, Vol. 48, No. 101297, April 2021;
Javier Villalba-Diez et al., “Integration of Quantum Simulation on a CNC Machine for In-
Process Control Visualization,” Sensors, Vol. 21, No. 15, July 2021.
31 Samir Khan et al., “On the Requirements of Digital Twin-Driven Autonomous

Maintenance,” Annual Reviews in Control, Vol. 50, No. 13-28, August 2020.
32 Tim van Erp and Bartłomiej Gładysz, “Quantum Technologies in Manufacturing

Systems: Perspectives for Application and Sustainable Development,” Procedia CIRP,

Vol. 101, No. 1120-1125, 2022.
33 Robert Liscouski, “How Quantum Computing Will Power the Future of Logistics,”

SupplyChainBrain, August 8, 2021.

Increasing data visibility has driven supply chains to shift from a relatively static
model (with infrequent data updates) to a more flexible model continuously updated
by real-time market supply and demand data. (See our Citi GPS report Global
Supply Chains: The Complicated Road Back to “Normal” for more details.)
Fundamentally, modern supply chains are immensely complex, as shown in Figure
17 below.

Figure 17. Supply Chain Dependency Analysis — Industrials

Source: Bloomberg, Citi Global Data Insights, Citi GPS

There are a number of promising applications of quantum computing in logistics,

such as traffic optimization to ease road congestion in real time. Quantum-optimized
supply chains could even bring environmental benefits, as transportation accounts
for 27% of all greenhouse gas emissions.34

We spoke to Dr. Alan Baratz, CEO of D-Wave, a quantum computing company

building both quantum annealing and gate-based quantum computers, about the
impact quantum computing may have on the manufacturing and logistics industry.

34United States Environmental Protection Agency, “Carbon Pollution from

Transportation,” accessed April 5, 2023.

Expert Interview with Dr. Alan Baratz, CEO of D-Wave

Q: Why do you feel that quantum computing has potential in manufacturing
and logistics?

Dr. Baratz: Quantum computing is a technology that today is already demonstrating

business value for manufacturing and logistics, specifically around the many
complex optimization problems native in these industries. Quantum computing uses
quantum mechanical effects to solve hard problems more quickly, more efficiently,
or in some instances, for the first time. This means that quantum computing’s
primary value is in revenue generation and/or cost savings. With the dynamism of
Industry 4.0, leaders in manufacturing and logistics are looking for innovative new
ways to harness technology to deliver upon those benefits. And these leaders are
beginning to see the impact of quantum computing and quantum hybrid (which
combines both classical and quantum) today. Whether in autonomous vehicles used
in factory floor automation, paint shop scheduling, or bin packing, quantum
computing has far-reaching impact.
Dr. Alan Baratz
CEO, D-Wave
Q: Could you elaborate on some of the optimization problems that D-Wave is
looking at applying quantum computing to?

Dr. Baratz: D-Wave’s annealing quantum computers are designed specifically for
optimization. And recent research suggests that because of the pre-processing
overhead of gate-model quantum computers, annealing quantum computers will
always be the best option for complex optimization problems. There are other use
cases in material and drug discovery and manufacturing that will require gate-model
quantum computers. This is why we announced last year that we are the only
company building both annealing and gate-model quantum computers. Think about
the value to manufacturing: longer-term, new metamaterials will be designed with
gate-model systems, while today and tomorrow factory automation improvements,
product customization, and optimized supply chains will deliver both existing and
new products to market more efficiently using quantum annealing.

Q: What do you see as the biggest barrier facing quantum adoption in

manufacturing and logistics?

Dr. Baratz: The biggest barrier, until recently, was the size of the systems. At D-
Wave, we’ve not only commercialized a 5000+ qubit system, but we’ve also brought
quantum hybrid computing to market. By using quantum and classical approaches,
business can now run problems up to 1 million variables. This unlocks an entirely
new set of potential use cases in manufacturing and logistics.

Artificial Intelligence
AI Has Become Ubiquitous, but Is Still a Half-Finished Technology
Artificial intelligence (AI) has been adopted in a variety of sectors to deliver
solutions and improve efficiency, but it is still a half-finished technology. The ultimate
form of AI, artificial general intelligence, should in principle be able to learn
anything.35 The capabilities of AI, however, are largely restricted by the computing
power currently available. With the lowest-priced GPU cloud on the market, it would
take 355 years and an electricity bill of around $4.6 million to train GPT-3 (an AI-
based language model) to produce human-like text.36 With Moore’s Law under
pressure in recent years, training larger language models will likely become an even
more difficult task for classical computers.

We investigate how quantum computing will potentially be able to enhance AI

capabilities through two examples: natural language processing and artificial neural
networking.

Figure 18. Exponential Growth of the Number of Parameters in NLP Models

Source: Microsoft Research

35 Peter Voss, Essentials of General Intelligence: The Direct Path to Artificial General

Intelligence (Berlin: Springer, 2007), 131-157.

36 Chuan Li, “OpenAI's GPT-3 Language Model: A Technical Overview,” Lambda Labs,

June 3, 2020.

Natural Language Processing

To achieve artificial general intelligence, machines must be able to understand and
respond to humans. However, equipping them with advanced linguistic systems is
difficult, as human language is filled with ambiguities and irregularities like
homophones, sarcasm, idioms, and metaphors.37 Natural language processing
provides a way to overcome this challenge through machine learning. It enables
computer systems to translate text, provide more relevant research results, and
even talk to us. Most people have probably already interacted with natural language
processing applications like Amazon’s Alexa, Apple’s Siri, Google’s voice input
search, or ChatGPT on Microsoft’s Bing platform.

Impressive as state-of-the-art natural language processing systems like ChatGPT-3

and ChatGPT-4 are, we are still far from achieving artificial general intelligence.
Natural language processing algorithms can only understand individual words, and
the meanings of sentences and paragraphs have to be inferred through some kind
of “black box.”

According to Bob Coecke, Chief Scientist at Quantinuum, AI could become

“meaning-aware” for the first time through quantum computing. This means that
instead of the risk of having everything taking place inside a black box, the flows of
meanings in quantum natural language processing are more clearly exposed and
understood. This is a completely novel approach that leverages the probabilistic
nature of quantum computing to better resemble how human language works.38

At present, quantum natural language processing research is still largely at the

stage of exploring and establishing theoretical foundations, and it will likely take
several years before it becomes ready for wide adoption.

Artificial Neural Networks

Enabling AI to recognize patterns like images and correlations is another challenge
to achieving artificial general intelligence. One of the best-performing machine
learning models for this is the artificial neural network, an algorithmic mimicry of the
biological neural network in our brains.

Figure 19. Layers of Biological Neurons (Left) and Layers of Artificial Neurons (Right)

Source: Citi GPS

37IBM, “What Is Natural Language Processing (NLP)?,” accessed April 5, 2023.

38Bob Coecke et al., Foundations for Near-Term Quantum Natural Language
Processing, Cambridge Quantum Computing Ltd. and Oxford University Department of
Computer Science, December 8, 2020.

Demand for more human-like AI is driving up the size of training data sets. IDC
predicts that worldwide data will grow from 33 ZB (ZettaBytes) in 2018 to 175 ZB by
2025.39 Eventually, even supercomputers are likely to be overwhelmed with the
amount of data generated.40 It has been proposed that fault-tolerant quantum
computers will in theory be capable of solving certain machine learning problems,
such as those solved by artificial neural networks, faster than classical computers.41

We spoke to Rob Hays, CEO at Atom Computing, who explained:

“It's been 10 years since the first paper was published on using GPUs for ‘ImageNet
Classification with Deep Convolutional Neural Networks.’ While that wasn’t the first
time a GPU was used in machine learning, the combination of highly-parallel GPU
hardware + neural network software architecture + a compelling use case in image
recognition marked an inflection point in the growth of AI. Quantum computers have
the potential to offer a similar inflection point in AI computing performance. Working
in conjunction with classical CPU + GPU systems, the inherent nature of quantum
computing systems will allow AI applications to take in more complex environmental
factors to simulate chemical and sub-atomic physical systems with a speed and
cost that isn’t practical today. This will unlock layers of complexity and natural world
simulations that aren't possible or practical today. Expanding the physical world
accuracy of AI will enable applications for improved materials development, medical
applications, complex system optimization, and others.”

Similar to quantum natural language processing, quantum neural networking is still

in its infancy. Many proposals and ideas have been forwarded regarding the optimal
structure and training of a quantum neural network to best utilize a quantum
computer’s potential.42 As we continue to scale on the hardware side, quantum
neural networking is likely to become more relevant.

We spoke to Jack Hidary, CEO of SandboxAQ, an enterprise SaaS company

looking at the convergence of AI and quantum technologies, about the impact
quantum computing may have on AI.

39 David Reinsel, John Gantz, and John Rydning, Data Age 2025: The Evolution of Data

to Life-Critical: Don’t Focus on Big Data, Focus on the Data That’s Big, IDC (sponsored
by Seagate), April 2017.
40 Avinash Chalumuri, Raghavendra Kune, and B.S. Manoj, “Training an Artificial Neural

Network Using Qubits as Artificial Neurons,” Procedia Computer Science, Vol. 171, No.
568-575, 2020.
41 Jacob Biamonte et al., “Quantum Machine Learning,” downloaded from arXiv, PDF,

May 10, 2018; Srinivasan Arunachalam and Ronald de Wolf, “A Survey of Quantum
Learning Theory,” downloaded from arXiv, PDF, July 28, 2017.
42 Stefano Mangini et al., “Quantum Computing Models for Artificial Neural Networks,”

downloaded from arXiv, PDF, May 20, 2021; Bob Ricks and Dan Ventura, “Training a
Quantum Neural Network,” Brigham Young University Department of Computer Science,
accessed April 5, 2023.

Expert Interview with Jack Hidary, CEO of SandboxAQ

Q: Why do you feel that quantum computing has potential in artificial
intelligence?

Jack: Quantum and AI are synergistic technologies. AI is at its best when we can
train it on large amounts of data, but when we look at a novel drug candidate, for
example, there is no big data. In these cases, we build up a new data set using the
quantum equations governing atomic interactions and then apply AI to that new data
to optimize for the best molecular structure to hit that condition.

Q: Could you elaborate on some challenges within artificial intelligence that

SandboxAQ is looking at applying quantum computing to?

Jack: One area where AI and quantum approaches are having great success is in
drug discovery – developing treatments for conditions such as brain cancer or
Alzheimer’s disease, two conditions which have stymied researchers for decades.

Jack Hidary When you develop a novel drug, there is little data available, so AI by itself is not as
CEO, SandboxAQ effective. With large-scale quantum simulation, we can examine the interactions
between chemical compounds and human receptors at the molecular level millions
of times in silico, and optimize the chemical structure using AI. This greatly
accelerates drug development, lowers R&D costs and reduces risk as the drug
candidate enters clinical trials.

Q: What do you see as the biggest barrier facing quantum adoption in

artificial intelligence?

Jack: More and more powerful GPUs are driving the advances we see in AI today.
Chips from Nvidia, Alphabet and other companies are giving us unprecedented
ability to build larger and larger AI models. However, as the chip industry faces
increased challenges in developing chips with even higher density of transistors due
to quantum limits, quantum computing will enter the picture. We foresee hybridized
classical-quantum computing as a way forward in creating more capable AI models.
The field of Quantum Machine Learning (QML) is at an early phase, but picking up
momentum. As quantum computers scale further and we can implement error
correction on qubits, we will see QML become a powerful tool in the AI toolchest.

Healthcare
A Potential Big Leap Forward
For the healthcare industry — long beset by inefficiencies and bottlenecks —
quantum computing could offer a big leap forward. The reason: with their
anticipated advantages in optimization, machine learning, and molecular simulation,
quantum computers are expected to provide the computational toolkit required to
solve some of healthcare’s most entrenched problems. As discussed below, this has
the potential to bring changes to the twin pillars of healthcare: drug discovery and
medical services.

Drug Discovery
Drug research and development are essentially chemistry problems characterized
by high failure rates, high R&D costs, and long development cycles. As the next
generation of computing, quantum computing could become the main driver for new
drug discovery. This is because drug molecules are quantum mechanical systems
themselves, meaning that quantum computers are inherently more suitable to
predict and simulate them than classical computers. More specifically, quantum
computing may help:43

 Model three-dimensional (3D) protein-folding structures in structure-based drug

discovery.

 Assess the synthesizability of a drug candidate more effectively in de novo

discovery.

 Predict the interaction of a drug candidate with multiple biological targets to

provide clues into toxicity, pharmacokinetics, and multitarget action.

These improvements could enable more effective and flexible computer-assisted

drug discovery tools, and thus reduce failure rates and costs. In the long run,
quantum computers may be able to expand current drug candidate libraries to
include macromolecules with complex 3D folding structures like proteins rather than
just small molecules. With direct hypothesis testing that helps better understand the
relationships between overall molecule structures and their specific medical
properties, quantum computers could possibly bring a paradigm-shifting change in
drug R&D — ultimately, introducing a new era of simulation-based drug discovery.

Medical Services
The amount of data collected by companies that provide medical services is
exploding. Healthcare data is expected to grow at a compound annual growth rate
(CAGR) of 36% through to 2025.44 The rapid increase in data volume is a result of
advancements in medical imaging and the increasing availability of real-time data
from continuous health monitoring.

43 Yuan Cao, Jacquiline Romero, and Alan Aspuru-Guzik, “Potential of Quantum

Computing for Drug Discovery,” IBM Journal of Research and Development, Vol. 62, No.
6, November 2018.
44 David Reinsel, John Gantz, and John Rydning, Data Age 2025: The Digitization of the

World: From Edge to Core, IDC (sponsored by Seagate), November 2018; Jessica Kent,
“Big Data to See Explosive Growth, Challenging Healthcare Organizations,” Health IT
Analytics, December 3, 2018.

Figure 20. 2018-25 Data Growth CAGR Across Various Industries

Media and Entertainment 25%

Financial Services 26%

Manufacturing 30%

Healthcare 36%

Global Datasphere 27%

0% 5% 10% 15% 20% 25% 30% 35% 40%

Source: IDC

On the one hand, big data is a powerful tool if properly used. On the other hand,
such an explosion of data also raises tough challenges for current data
management systems. One recent report estimated that 99% of the data we
generate is not analyzed, and of the 1% that is, this is mostly done in a discrete and
shallow manner.45

Quantum computing could help address this challenge and lead to:

 Better Diagnoses: Care providers may be able to make earlier, more accurate,
and faster diagnoses — ultimately saving lives.46 With improved data processing
capabilities, care providers may be able to continuously monitor more physical
indicators for our health and raise the alarm much earlier when something goes
wrong. This could enable preventative analysis to become more predictive,
helping providers determine in advance which procedure should be performed at
what time.47 Furthermore, quantum computers may even bring diagnostic
procedures down to a cellular level, and for example, be used to detect cancer
cells more accurately.48

45 Nicolaus Henke, Ari Libarikian, and Bill Wiseman, “Straight Talk About Big Data,”
McKinsey Quarterly, October 28, 2016.
46 Rishabha Malviya and Sonali Sundram, “Exploring Potential of Quantum Computing in

Creating Smart Healthcare,” The Open Biology Journal, Vol. 9, No. 56-57, September
2021.
47 Devansh Mehta, “Quantum Computing Will Completely Change the Healthcare

Infrastructure to New Level and Help Transform the Healthcare From Preventive To
Predictive Healthcare: A Future Perspective,” Journal of Bio Innovation, Vol. 9, No. 6,
November 2020.
48 Nilima Mishra et al., “Cancer Detection Using Quantum Neural Networks: A

Demonstration on a Quantum Computer,” downloaded from arXiv, PDF, November 1,

2019.

 Tailored Treatment: In the future, quantum computers could potentially enable a

better and more detailed understanding of personal illness and deliver precision
medicine that accounts for the patient’s age, race, gender, and genetic makeup,
among other factors.49 This would be an improvement over the current umbrella
approach, whereby diagnosis and treatment are based heavily on patient-
reported symptoms. As such, the current approach fails to provide personalized
medical actions at the individual level.50 This is important because current
medical care itself only contributes 10%-20% towards final patient outcomes
today, while health-related behaviors, socioeconomic factors, and environmental
aspects account for the rest.51

Figure 21. Umbrella Diagnosis vs. Precision-Based Diagnosis

Source: IBM

 Enhanced Insurance Risk Assessments and Fraud Detection: Quantum

computers are expected to enable more granular modeling for insurance
premium calculations, significantly improving the accuracy of individual health
plan pricing, as well as likely reducing premiums.52 In addition, the anticipated
superior machine learning capacity of quantum computers is expected to help
spot fraudulent medical claims and abnormal behavior accurately, lowering
premiums further.53

We spoke to Peter Chapman, CEO of IonQ, which builds quantum computers

based on trapped ion qubit technology, about the impact quantum computing may
have on the healthcare industry.

49 Rishabha Malviya and Sonali Sundram, “Exploring Potential of Quantum Computing in

Creating Smart Healthcare,” The Open Biology Journal, Vol. 9, No. 56-57, September
2021.
50 IBM, Exploring Quantum Computing Use Cases for Healthcare, June 2020.

51 Carlyn M. Hood et al. “Relationships Between Determinant Factors and Health

Outcomes,” American Journal of Preventive Health Medicine, Vol. 50, No. 2, February
2016.
52 IBM, Exploring Quantum Computing Use Cases for Healthcare, June 2020.
53 Ibid.

Expert Interview with Peter Chapman, CEO of IonQ

Q: Why do you feel that quantum computing has potential in healthcare?

Peter: Quantum computers were originally envisioned to help solve hard chemistry
problems — and there are few harder than understanding the human body and its
many systems. As quantum computers become more accessible and more
powerful, healthcare providers start to envision tackling increasingly ambitious
problems from a genuine understanding of our bodies’ biochemistry.

Whether it’s simulating complex compounds for a new drug, identifying more
effective therapies, or gaining an overall better understanding of how one treatment
impacts another, quantum computers offer the greatest potential of upending the
status quo and delivering more personalized, effective care to patients around the
world.

Q: Could you elaborate on some challenges within healthcare that IonQ is

looking at applying quantum computing to?
Peter Chapman
CEO, IonQ Peter: On average, it takes several years and billions of dollars to take a new drug
from the lab to the marketplace, with years of R&D effort required before a
candidate molecule even makes it into clinical trials. Quantum computing has the
potential to expedite the entire process by simulating chemical reactions via
software instead of in the lab.

In drug development, for example, scientists must first identify molecules with a set
of desired properties, such as the ability to bind with specific proteins. While much
of this work is already computer-assisted, extensive lab work is still required to fully
understand candidate molecules, with many cycles of synthesis and testing required
to find a suitable candidate. Quantum computers could quickly screen billions of
molecules to identify suitable candidates for clinical trials, skipping the process of
identifying what works or doesn’t work earlier on, improving the cost and speed of
development, increasing the likelihood of success, and improving the most
important thing of all: patient outcomes.

Energy and Climate

The Ability to Reduce Carbon Emissions
The energy and climate sectors are well poised to take advantage of quantum
computing’s future capabilities, especially in the areas of molecular simulation and
optimization. One recent report claims that quantum-based climate technologies
could help us avoid 7 gigatons of carbon emissions every year by 2035. This
equates to around 20% of global carbon emissions today, and thus could contribute
a great deal to achieving net-zero targets.54

Energy
Although the world has seen increased adoption of renewable energy, fossil fuels
are still the primary source of the world’s energy use, contributing to over 60% of
global energy generation for the past decade.55

Figure 22. Global Energy Generation by Scenario

In the future, quantum computing could benefit various energy sectors in different
ways, including:

 Oil: Improving identification of potential crude oil reservoirs by simulating the

interactions of crude oil, water, and gas molecules with the surface of rocks.

 Green Hydrogen: Developing new catalysts for water electrolysis, which could
make green hydrogen economically viable and encourage wider adoption. One
report estimated that quantum computing has the potential to bring down the cost
of green hydrogen by 60% (from $3/kg-$6.5/kg to $1.2/kg-$2.6/kg).56

54 Peter Cooper, Philipp Ernst, Dieter Kiewell, and Dickon Pinner, “Quantum Computing

Just Might Save the Planet,” McKinsey, May 19, 2022.

55 International Energy Agency (IEA), World Energy Outlook, November 2019.

56 Peter Cooper, Philipp Ernst, Dieter Kiewell, and Dickon Pinner, “Quantum Computing

Just Might Save the Planet,” McKinsey, May 19, 2022; Tom DiChristopher, “Experts
Explain Why Green Hydrogen Costs Have Fallen and Will Keep Falling,” S&P Global,
March 5, 2021.

 Energy Storage: Simulating the complicated chemistry of prototype battery

designs and accelerating the development process. Durable and efficient battery
technology is essential to the large-scale adoption of renewables, as most
sources of renewable energy cannot continuously generate electricity. However,
battery technology is only incrementally improving nowadays, with an
approximate 5% improvement in efficiency every year.57

Agriculture
Nearly all fertilizers that help feed us today are made from ammonia, and about
50% of the world’s food production relies on these fertilizers.58 However, the most
commonly utilized ammonia production process today, the Haber-Bosch process,
dates back to the 1910s and requires stringent conditions, including very high
pressures and temperatures. The ability to produce ammonia in a more efficient and
environmentally friendly way could result in both cheaper and less energy-intensive
fertilizers.

As quantum computers’ anticipated ability to simulate the quantum properties of

molecules increases, it may eventually become possible to identify and produce the
enzyme that catalyzes ammonia synthetization at a large scale.59 Thus far,
scientists have been unable to reproduce this enzyme, as the chemical composition
of soil is particularly complicated, and the current computational power of classical
computers is insufficient to run the simulations needed to determine the enzyme.

Climate Change
Quantum computers could also have an impact in carbon capture, utilization, and
storage (CCUS), which, according to the International Energy Authority (IEA), is
expected to make up 50% of heavy industry’s emission reductions by 2050.60

57 Scott K. Johnson, “Eternally Five Years Away? No, Batteries Are Improving Under

Your Nose,” Ars Technica, May 24, 2021.

58 Leigh Krietsch Boerner, “Industrial Ammonia Production Emits More CO2 Than Any

Other Chemical-Making Reaction. Chemists Want to Change That,” Chemical &

Engineering News, June 15, 2019.
59 Sebastian Jeon, “Feeding the World with Die Rolls: Potential Applications of Quantum

Computing,” Dartmouth Undergraduate Journal of Science, Vol. 20, No. 1, 2017.

60 Amy Flower, “How Quantum Computing Can Help Tackle Climate Change,” Riverlane,

November 4, 2021.

Figure 23. The CCUS Value Chain

Today, most CO2 reduction processes are quite expensive, in part due to the
involvement of precious metals. In the best-case scenario, one gram of a metal
catalyst could capture 100 grams of carbon-dioxide emissions.61 This means that a
1% reduction in global emission would require 3.63 million metric tons of this metal
catalyst.62 Quantum computers could potentially leverage their advantage in
simulation to help discover new catalysts for carbon capture to absorb carbon
directly out of the air more efficiently. The captured CO2 could then be used in the
production of synthetic hydrocarbons, polymers, and building materials.63 Such
work is already taking place — for example, in 2020, one energy company already
announced that it would be working with the company now known as Quantinuum to
improve materials for CO2 capture.

The simulation capabilities of quantum computing could also help us better

understand the Earth’s climate system and ultimately allow us to react to natural
disasters more efficiently and accurately.64 For example, by combining quantum and
classical computing, Rigetti claims to have developed an effective way to produce
high-quality synthetic weather radar data.65

We spoke to Rajeeb Hazra, CEO of Quantinuum, about the impact quantum

computing may have on the energy and climate sectors.

61 Katie Lamb, “Carbon Capture and Conversion Must Not Rely on Rare Metals,”

Phys.org, January 29, 2019.

62 Ibid; IEA, “Global CO2 Emissions Rebounded to Their Highest Level in History in

2021,” March 8, 2022.

63 Alcimed, “CCUS technology: CO2 As a Resource or a Waste?,” March 12, 2020.

64 Manmeet Singh et al, “Quantum Artificial Intelligence for the Science of Climate

Change,” downloaded from arXiv, PDF, accessed April 6, 2023.

65 Rigetti, Rigetti Enhances Predictive Weather Modeling with Quantum Machine

Learning,” December 1, 2021.

Expert Interview with Rajeeb Hazra, CEO of Quantinuum

Q: Why do you feel that quantum computing has potential in the energy and
climate sectors?

Rajeeb: Quantum computing is a profoundly new and different technology, drawing

on the laws of quantum mechanics to tackle certain computational tasks that would
otherwise be intractable. The kinds of problems that quantum computing shows
promise in tackling include modelling complex molecular or material systems, or
optimizing industrial processes.

Q: Could you elaborate on some challenges within energy and climate that
Quantinuum is looking at applying quantum computing to?

Rajeeb: There are many examples within energy and climate change where
quantum computing appears to offer novel and extremely relevant solutions, for
example, in areas such as low- or no-carbon energy production, in manufacturing,
supply chain optimization and distribution, or researching scalable methods that
Rajeeb Hazra
could help to reverse historic atmospheric emissions. Quantum computing offers a
CEO, Quantinuum
path to the rapid and cost-effective development of new complex materials, which in
turn may have many uses for energy production and climate change mitigation.
Examples include the development of highly efficient photo-voltaic cells, or the
discovery of highly efficient chemical processes to enhance the efficiency and
performance of hydrogen fuel cells for electric vehicles (work we continue to do with
major industrial players in automotive and aerospace).

Q: What do you see as the biggest barrier facing quantum adoption in the
energy industry?

Rajeeb: Quantum computing is a complex and rapidly advancing technology. This

presents two related challenges, hardware development, and identifying suitable
use cases that could be amenable to a quantum speed-up. As we approach the
period where quantum computing hardware offers a meaningful advantage over
classical computers, large corporations will increasingly seek to work with full stack
companies like Quantinuum to accelerate their quantum know-how, build out their
quantum workflows, and position themselves to benefit from the first developments
that offer a quantum advantage.

Q: Is quantum computing going to be good for improving battery technology?

Rajeeb: The evidence is very positive in this field. Quantinuum has collaborated
with the DLR – the German national aeronautics and space research center – on a
broad project to investigate the use of quantum algorithms to model battery cells.
The project specifically looked at using Quantinuum’s quantum algorithms for
solving partial differential equations to render a simulation of a lithium-ion battery
cell, a central use-case in this field. We have also supported global enterprise
partners to explore today’s quantum devices for simulating chemical reactions
within both batteries and hydrogen fuel cells with our state-of-the-art quantum
chemistry platform, InQuanto.

Finance
Tackling Probabilities in a Data-Heavy Environment
Several financial services activities, from securities pricing to portfolio optimization,
require the ability to assess a range of potential outcomes. For this, banks rely on
algorithms and models to calculate statistical probabilities. However, today's data-
heavy environment has led to the need for ever-more powerful computers to
calculate probabilities accurately. Therefore, the opportunity for quantum computing
and its potential performance gap with classical computing methods keeps growing.

Quantum computing is still an emerging technology, and its use cases in finance
have been an active area of experimental research and development. While several
use cases are still in the proof-of-concept stage, we are seeing the emergence of a
few practical applications. We believe quantum computing in finance could open up
a range of compliance improvements, efficiencies, and new market opportunities.
We list below some of the promising future applications.

Targeting and Prediction

Quantum computing could help overcome the limitations of existing analytical
models to sift through large amounts of behavioral data. This may enable financial
institutions to offer more personalized products and services to the right customers
in real-time, resulting in greater revenue opportunities.

Quantum computers could also prove superior in finding patterns, performing

classifications, and making predictions that are often not possible today due to
complex and siloed data structures. Banks with large client databases could
particularly benefit from the use of improved searching algorithms.

Using quantum computers to optimize loan portfolios focusing on collateral could

also allow financial institutions to improve their offerings, possibly lowering interest
rates and freeing up capital. For instance, banks may be able to offer more targeted
products and services to their customers based on purchasing trends, preferences,
and demographics. This could also help banks selectively tap into the underbanked
small and medium-sized enterprise (SME) market.

Portfolio Optimization
An important aspect of the financial industry relates to the pricing of financial
instruments and estimating their risk. Regulators also require banks to perform
various stress tests in order to hold adequate capital depending on their risk-
weighted assets. Analytical models may often be too simplistic to capture the
complex dependencies between different elements, which may otherwise require
extensive computational power.

Quantum machine learning is expected to help enhance robo-advisory capabilities

and trading algorithms. It could also optimize collateral management across
exchanges and currencies, facilitate dynamic arbitrage (for example, in cross-
currency debt or cryptocurrency markets), and enable combinatorial and convex
portfolio optimization methods.

Improving computation speed in certain tasks could also reduce complexity in

traditional environments, including settlement processes, associated capital
requirements, systemic risk, and operational costs. Quantum computing could help
optimize trade settlement from once-a-day cycles to more frequent intraday cycles.
Many banks are exploring the use of quantum computing for trading strategies,
portfolio optimization, asset pricing, and risk analysis.

Risk Management and Fraud Detection

Risk management plays a central role in the financial sector and includes aspects
such as credit risk, liquidity risk, and market risk, which are often estimated using
models. Quantum computers are expected to help build models incorporating a
large number of variables to create comprehensive risk profiles and facilitate better
decision making, as well as providing a significant increase in speed over
established classical algorithms.

The use of quantum computing is expected to improve the performance of Monte

Carlo-based options pricing and valuation methods. Similarly, it will likely be able to
help in the modeling of complex financial market trading activities (e.g., valuation
adjustment models for derivatives) wherein different scenarios are applied to a
portfolio of financial instruments to calculate its value at a time in the future.

Figure 24. Potential Benefit to Financial Services Activity from Quantum Computing

Source: IBM

That said, quantum computers may also pose risks to the finance industry. Financial
institutions are privy to sensitive customer data, which is often encrypted using
techniques that even a modern supercomputer would be unable to decrypt (or
“break”) without an extremely long processing time — often in the order of
thousands or millions of years. As we discuss further in the “Cybersecurity” section,
calculations show that a sufficiently advanced quantum computer could break
today’s encryption standards in just a few hours, thus posing a risk to the global
financial ecosystem.

Furthermore, as today’s cryptography standards are a key building block to the

blockchain and other distributed ledge technologies, quantum computers also pose
a risk to the cryptocurrency ecosystem and the elements of the financial industry
being built on top of it — which we discuss in the “Cryptocurrency” section.

We spoke to Matt Johnson, CEO of QC Ware, which provides professional services

and enterprise software for quantum computers, about the potential impact quantum
computing may have on the finance industry.

Expert Interview with Matt Johnson, CEO of QC Ware

Q: Could you elaborate on some challenges within finance that QC Ware is
looking to apply quantum computing to?

Matt: Intra-day derivative pricing and performance analysis for complex portfolios is
something that we believe will be a highly sought-after solution once adequate
hardware becomes available. Currently, these pricing calculations happen overnight
and take a few hours to complete. The results are provided to the traders to inform
their decisions during the trading day, but the information can get stale very quickly,
especially on days with significant market volatility. Being able to provide traders
with a process that can price a very complex portfolio in minutes rather than hours
will be a significant game-changer in the finance sector, and that is what QC Ware
is currently working on.

Q: Do institutions need to hire their own teams of quantum algorithm experts

in order to integrate quantum computing into their existing finance
Matt Johnson
infrastructure?
CEO, QC Ware
Matt: Industry leaders will have to hire their own experts to maintain an in-house
point of view on developments and to be able to validate what software and
hardware vendors are proposing. We believe that 80% of quantum computing
solutions that will be utilized in the market will come in the form of pre-packaged,
off-the-shelf applications. However, we also believe that industry leaders will always
look to innovate in-house and build unique and custom approaches that provide
specific advantages.

Q: What do you see as the biggest barrier facing quantum adoption in

finance?

Matt: The barrier is the same as for any other software solution. It needs to provide
clear value, it must be easily accessible, and it should be integrated with existing
technology and processes.

Q: Is quantum computing going to be good for high-frequency trading?

Matt: It is very hard to see something like this happening within the next 20 years.
Our best guess is that in order to be able to beat classical solutions in high-
frequency trading, the quantum computing ecosystem would need to develop a
purely quantum pipeline starting with the input data that would have to live in some
quantum state, the communication pipelines, the computation of the recommended
trade, and its execution.

Cybersecurity
Opportunities rarely come without risks. One serious risk posed by quantum
computing is to data privacy and cybersecurity. The power of quantum computing
could be misused for breaking encryption, which is at the heart of all forms of
electronic communication and data storage across the world today. This includes
digital authentication, public key infrastructure (PKI), and mobile chat systems.

How Data Is Currently Protected

There are mainly two types of data encryption methods used today. One is
symmetric-key encryption, where the sender and receiver have identical
(symmetric) digital keys to encrypt and decrypt data. The other is public-key
(asymmetric) encryption, where a publicly available key encrypts messages for a
recipient who has a carefully guarded private key for decryption.

Figure 25. Symmetric vs. Asymmetric Encryption

Source: Citi GPS

The main difference is that symmetric encryption requires the use of shared secret
keys; otherwise, it is faster and believed to be more secure. Public key encryption,
on the other hand, has the advantage of only requiring an authentic copy of publicly
available keys (i.e., keys that need not be kept secret), and it provides a means for
establishing symmetric keys for the more efficient symmetric encryption algorithms.
However, public key encryption is generally believed to be more susceptible to
cryptanalysis (the process of identifying and exploiting weaknesses in cryptographic
algorithms without knowing the secret key). This is because it requires additional
mathematical structure compared to symmetric key encryption algorithms.

RSA encryption is one of the most widely used public-key cryptographic standards.
It is constructed in a way that is easy to decrypt using one’s private key, but
otherwise extremely difficult for anyone else. RSA encryption relies on the fact that,
while a classical computer can easily multiply large prime numbers together, the
reverse process of factoring (determining which large prime numbers multiply
together to get to the final number) is extremely difficult. This is because while there
exists a highly efficient algorithm for multiplying large numbers (the long
multiplication method we all learn in school), there is currently no known efficient
algorithm that can run on a classical computer to reverse this process.

Figure 26. Factorization vs. Multiplication

Source: Citi GPS

While a smartphone or laptop can easily multiply large prime numbers, even a
classical supercomputer is unable to reverse such a calculation without an
extremely long processing time — often in the order of thousands or millions of
years, depending on the length of the number. Leveraging this impracticality for
classical computers to factor large numbers, current public-key encryption has
become uncrackable by using very long key pairs — such as 2048-bit keys, which
equate to 617-digit decimal numbers.

The Threat to Encryption Standards

This foundation to asymmetric encryption was shaken by Shor’s algorithm,
proposed by Professor Peter Shor in 1994.66 Shor’s algorithm theorized that a large
fault-tolerant quantum computer could find the prime factors of large numbers in a
fraction of the time required by classical computers. While the nuances of the actual
algorithm are far beyond the scope of this report, the broader point is that Shor’s
algorithm could provide a means to any hostile actor with a large enough quantum
computer to start with a public key of an RSA public-key cryptosystem and reverse-
engineer the associated private key, leaving asymmetric encryption techniques like
RSA vulnerable to quantum attack.67

This is because encryption like RSA relies on the principle that the difficulty of
factoring large numbers scales exponentially for classical computers, whereas a
successful implementation of Shor’s algorithm changes this to scale in a polynomial
fashion, as Figure 27 shows. In fact, it has been calculated that sufficiently
advanced quantum computers could decrypt even 4096-bit key pairs in just a few
hours using Shor’s algorithm. The effect of this would be to potentially render
communications as insecure as if they were not encrypted at all.

Figure 27. Quantum Computing's Potential for Significant Speed-up Over Classical
Computers
Type of Scaling Time to Solve Problem
Classical algorithm with 10 2 330 3,300 Age of the
exponential runtime secs mins years years universe
Quantum algorithm with 1 2 10 11 ~24
polynomial runtime min mins mins mins mins
Source: IBM

66 Peter Shor, "Algorithms for Quantum Computation: Discrete Logarithms and

Factoring," Proceedings 35th Annual Symposium on Foundations of Computer Science,

124-134, November 1994.
67 Dorothy E. Denning, “Is Quantum Computing a Cybersecurity Threat?,” American

Scientist, March 2019.

As for symmetric encryption: While it is not impacted by Shor’s algorithm, it is

affected by Grover’s algorithm. Grover’s algorithm is a quantum computing
algorithm proposed by Lov Grover in 1996 that makes it far simpler to search
through unstructured data sets. Within the context of cybersecurity, it has been
found that a large fault-tolerant quantum computer running Grover’s algorithm could
provide a quadratic speed-up in attacking symmetric encryption standards like
Advanced Encryption Standard (AES).68 Consequently, to resist such attacks and
maintain the same level of security strength in the quantum era, it would be
necessary to double key sizes. For AES, this means using 256-bit keys instead of
today’s 128-bit keys.69

Quantum computers currently have too little computing power and are too error-
prone to crack today’s strong codes. For context, one estimate by experts suggests
a quantum computer would need at least 70 million physical qubits to crack 2048-bit
RSA encryption.70 Considering IBM’s 433-qubit Osprey is the largest quantum
computer we know about today, this would, on the face of it, suggest it could be a
long time before we reach that point. In addition to needing almost 100,000 times
more qubits than today’s largest quantum computers, any fault-tolerant quantum
computers that could be considered useful in cracking today’s encryption standards
would also need an error rate that is 1/100th of what today’s best quantum
computers can reach.71 However, other factors mean that we may not be as far
away from the day when the quantum threat materializes as many think.

This is because things are moving faster in this area than is generally understood,
and technologies are developing in various directions. Not only are we seeing
quantum computers with more qubits every year, but new algorithms are reducing
the computing power and error-correction rates needed for quantum computers to
crack current day encryption by orders of magnitude. For instance, Google and the
KTH Royal Institute of Technology in Sweden claimed in 2019 that they found a
more efficient algorithm for quantum computers to break encryption standards.
Using their new algorithm, in theory, even a 20 million-qubit computer could break
2048-bit encryption in a mere 8 hours.72

To gauge how far away this quantum threat is, the Global Risk Institute interviewed
40 of the world’s leading experts, asking when they thought quantum computers
would be able to break the 2048-bit RSA encryption used today within 24 hours. A
small share (2.5%) of the experts interviewed indicated there was a 50% or more
likelihood of this occurring in the next 5 years. However, this rose to 22.5% and 55%
when asked about the same likelihood of this occurring in the next 10 or 15 years,
respectively. In fact, the vast majority (92.5%) of the experts estimated there was a
50% or more likelihood of quantum computers being able to break 2048-bit RSA
encryption in the next 20 years.73

68 Sandeep Rao et al., “The AES-256 Cryptosystem Resists Quantum Attacks,”

International Journal of Advanced Computer Research, Vol. 8, No. 3, April 2017.
69 Ibid.
70 QuantumXChange, “What Is The Impact Of Quantum Computing On Cybersecurity?,”
accessed March 7, 2023.
71 Emily Grumbling and Mark Horowitz, Quantum Computing: Progress and Prospects

(Washington, DC: The National Academies Press, 2019).

72 MIT Technology Review, “How A Quantum Computer Could Break 2048-Bit RSA

Encryption In 8 Hours,” May 30, 2019.

73 Dr. Michele Mosca and Dr. Marco Piani, Quantum Threat Timeline Report 2022,

Global Risk Institute, December 2022.

It is important to note that such a stringent timeframe of 24 hours to break

encryption inherently provides a very conservative estimate as to when the quantum
computers would pose a threat. The quantum threat to cybersecurity is still very
real, even if, rather than the 24 hours posed in the question above, quantum
computers are able to break such modern encryption techniques in several days,
weeks, or even months.

The Scope of the Problem

If these encryption methods are broken, nearly all encrypted communications and
data could be transparent to hostile actors with access to a quantum computer. It is
not difficult to imagine the scope of potential consequences if a nefarious group
were to access and steal confidential information from governments, such as state
secrets.74 Similarly, the interconnected devices that make up Industry 4.0 are also at
high risk from quantum attacks. With commercial machinery becoming increasingly
interconnected under IoT systems, quantum cybercriminals could in theory attack
and gain control over physical assets.

The risk that quantum computers pose to data security may be even more pressing
than one would think: Hostile actors may be able to start harvesting encrypted data
now, with the intent of utilizing a quantum computer to decrypt it in the future —
known as a “Harvest Now, Decrypt Later” (HNDL) attack. This is a particularly
important issue for companies and institutions that manage highly sensitive data,
such as pharmaceutical companies, large banks, governments, and intelligence
agencies. In practice, the use of HNDL attacks means a large amount of stolen data
may be currently residing in databases, ready to be decrypted and exposed in the
coming years.

Ultimately, what this means is that the post-quantum security threat exists in the
present, not the future. Financial institutions and nearly every industry that deals
with sensitive data need to review their security strategies and take appropriate
steps, including creating multiple data classifications, each with their own encryption
algorithm and encryption key, to protect against the quantum threat.

Post-Quantum Cryptography
As well as the risk to national security, breaches of current cryptographic systems
by quantum computers could carry potentially significant consequences for civilian
communications and the confidentiality of corporate information. Consequently,
there are many endeavors by both governments and private companies to develop
a quantum-resistant form of cryptography — commonly referred to as Post-
Quantum Cryptography (PQC).

A quantum-resistant form of cryptography is a cryptographic problem that is

intractable (i.e., very difficult) for both classical and quantum computers to solve.
Fortunately, such cryptographic problems are not difficult to define, and can be
retrospectively implemented into classical hardware. However, on the other hand,
the actual commercial transition process to PQC will likely be quite challenging, as
this will require the mass adoption of a completely new standard. Organizations
around the world have been trying to determine what this transition might look like in
order to recommend best practices.

74 Bitdefender, “How Quantum Computing Will Impact Cybersecurity,” March 13, 2021.

The U.S. National Institute of Standards and Technology (NIST) published the first
public encryption standard, DES, in 1977.75 Having since developed a full portfolio
of classical encryption standards, NIST began the search for appropriate PQC
standards in 2016 when it called for proposals. 76 Eighty-two initial submissions
were received from 25 countries in 2017, with 69 making it to the second-round
selection stage. After several more selection rounds, NIST recently announced the
four final candidates for PQC algorithms: One (CRYSTALS-Kyber algorithm) is for
general encryption, while the other three (CRYSTALS-Dilithium, FALCON,
SPHINCS+) are specifically for digital signatures.77 According to the NIST’s latest
timeline, draft standards are currently planned for release by 2024.78

The U.S. government, specifically the National Security Agency (NSA), recently
notified all its National Security Systems (NSS) owners, operators, and vendors of
the future quantum-resistant algorithm requirements. The press release explicitly
stated that one of the reasons now was the time to plan, prepare, and budget for the
transition to quantum-resistant algorithms was because of foreign pursuits in
quantum computing. Rob Joyce, Director of NSA Cybersecurity, explained, “This
transition to quantum-resistant technology in our most critical systems will require
collaboration between government, National Security System owners and
operators, and industry.”79

Many cryptography practitioners expect all governments and institutions that

interact with them (which includes a huge number of corporations) will be mandated
to adopt the new standards. However, the migration process to a new PQC
standard is not something that could typically be done overnight — in the case of
governments, this process could take 5-10 years to complete, while for large
companies, the transition will probably take multiple years as well. At the same time,
it is important to note that PQC standards have not yet been fully established. As
the Director of NSA Cybersecurity said, “We want people to take note of these
requirements to plan and budget for the expected transition, but we don’t want to
get ahead of the standards process.”80

Figure 28. NIST PQC Standard Timeline

Year Milestone
2016 Criteria and requirements for PQC standard released and call for proposals
2017 1st round candidates announced
2018 1st PQC Standardization Conference
2019 2nd round candidates announced and 2nd PQC Standardization Conference
2020 3rd round candidates announced
2021 3rd PQC Standardization Conference
2022 4th round candidates announced and 4th PQC Standardization Conference
2022/2024 Draft standards available
Source: NIST, Citi GPS

75 National Institute of Standards and Technology (NIST), “Data Encryption Standard,”

PDF, January 15, 1977.
76 NIST, “Computer Security Resource Center,” accessed March 7, 2023.

77 NIST, “NIST Announces First Four Quantum-Resistant Cryptographic Algorithms,”

July 5, 2022.
78 NIST, “Post-Quantum Cryptography: Workshops and Timeline,” last updated July 17,

2023.
79 NSA, “NSA Releases Future Quantum-Resistant (QR) Algorithm Requirements for

National Security Systems,” September 7, 2022.

80 Ibid.

In the interim period, until the finalized PQC standards are published and adopted,
institutions are susceptible to the HNDL attacks we described above. One report
explained that the use of longer encryption key lengths, such as migrating from
1024-bit RSA encryption to 2048-bit RSA encryption, could provide an additional 1-3
years of protection against the threat posed by early prototypes of error-corrected
quantum computers.81 Furthermore, some experts we spoke to explained that a
number of companies are already moving to 2048-bit as their standard.

Notably, it is important to be aware that the adoption of the PQC standards does not
necessarily provide 100% protection against quantum computers. This is because,
as with any standard, there is always a chance that a new revolutionary algorithm
could be created to decrypt it, especially as large quantum computers become
available for codebreakers to experiment with. Hence, the notion of cryptographic
agility is key to an institution’s ability to address the future threat of quantum
computers. For instance, one form of cryptography that was considered “quantum-
safe” was recently hacked by a 10-year-old PC.82

We discuss this further in our “How to Prepare: For the Quantum Threat” section of
this report, where we look closer at the problem and provide some high-level
guidelines for corporates wishing to better understand the threat quantum
computers pose to the security of their data. Regardless, the quantum threat to
cybersecurity could take significant time and cost to address, similar to the Y2K
problem.

Lessons from The Y2K Problem

The threat quantum computing poses to cybersecurity is in many ways akin to the
Y2K problem at the turn of the millennium, when computers were incapable of
processing dates containing “00” after “99” due to some initial program design to
save memory. It is estimated that the U.S. alone spent around $100 billion in
preparation for the problem, with around $9 billion provided directly from federal
government.83 Globally, various estimates have been provided for Y2K-readiness
spending, from $200 billion to $850 billion.84 IDC even launched a specific project,
Project Magellan, back in 1999, with the sole purpose of studying the impact of the
Y2K bug — it estimated a $320 billion bill spend over seven years, with a peak
spending of $101 billion in 1999.85

The cost was not the only significant aspect of the Y2K problem. The scope of the
impact was also very significant — there were over 100 million computers in the
U.S. in 1999, and all of them were potentially at risk.86 This is all the more stark
when we consider that the Y2K problem occurred over two decades ago, when the
internet only had 248 million users, compared to the over 5 billion users it has
today.87 Furthermore, $1 in 1999 would have an equivalent purchasing power of
$1.81 today as a result of inflation.88

81 McKinsey, “When—and How—to Prepare for Post-Quantum Cryptography,” May 4,

2022.
82 Charles Q. Choi, “‘Quantum-Safe’ Crypto Hacked by 10-Year-Old PC,” IEE Spectrum,

August 19, 2022.

83 Farhad Manjoo, “Was Y2K a Waste?,” Slate, November 11, 2009.

84 IT Web, “Y2K Cost Beyond Reckoning, Global Coordinator Says,” January 4, 2000.

85 Jack Schofield, “Money We Spent,” The Guardian, January 4, 2000.

86 Infoplease, “U.S. Households With Computers and Internet Use, 1984–2014,” updated

June 26, 2019.

87 Internet World Stats, “Internet Growth Statistics,” accessed March 7, 2023.
88 In 2013 Dollars, “CPI Inflation Calculator,” accessed March 7, 2023.

The CEO of SandboxAQ, a SaaS provider combining AI and quantum technology to

deliver solutions, pointed out in a recent quantum technologies conference that over
20 billion devices are currently exposed to the decryption risk from quantum
computers (250 times more than the number for Y2K), and all of these will all need
to migrate to a PQC standard to make them resistant to quantum attacks.89

Unlike the Y2K problem, however, the quantum threat to cybersecurity is less well-
defined, in that we don’t know when it will be realized. Estimating a precise date is
made all the more complicated given that players are unlikely to disclose how close
they are to making a quantum computer capable of breaking current encryption
standards. Nonetheless, the ways that countries and corporations successfully
managed their Y2K problems may provide a blueprint for how they can address the
threat quantum computing poses to encryption. Irrespective of any such exact date,
due to the threat of the “Harvest Now, Decrypt Later” attacks described earlier, the
potential risk to data security could be happening right now.

We spoke to Dr. Michele Mosca, CEO and Co-Founder of evolutionQ, which

provides quantum-safe cybersecurity products and services, about when this
quantum threat to cybersecurity could occur and what organizations can do now.

89 Lionel Sujay Vailshery, “Internet of Things (IoT) and Non-IoT Active Device
Connections Worldwide from 2010 to 2025,” Statista, September 6, 2022; David Joseph
et al., “Transitioning Organizations to Post-Quantum Cryptography,” Nature, Vol 605,
May 11, 2022.

Expert Interview with Dr. Michele Mosca, CEO and Co-

Founder of evolutionQ
Q: When are we likely to see the “quantum threat” materialize?

Dr. Mosca: By the time we know for certain, it’s too late to respond. It’s a matter of
aligning your risk appetite or tolerance to the likelihood of the threat in five years or
in 10 years, and so on. And for very critical systems where the risk tolerance is low,
the latest Global Risk Institute report indicates that even the five-year timeline is
already a concern.

Q: What can organizations start doing now to prepare for the quantum threat?

Dr. Mosca: Organizations must position themselves to manage this new threat
using technology lifecycle management and not crisis management. The journey to
a quantum-safe posture for your organization can be broken up into six phases:
preparation, discovery, risk assessment, risk mitigation, migration, and validation.
evolutionQ pioneered the “Quantum Risk Assessment” which brings organizations
through the first three stages toward quantum-readiness, and positions them to
Dr. Michele Mosca
prioritize their activities for the last three stages. There are great tools, both open-
CEO and Co-Founder, evolutionQ
source and proprietary, to support the migration to quantum-safe cryptography,
including our Quantum Delivery Network product.

Q: What realistic role does Quantum Key Distribution (QKD) play in protecting
against quantum attacks? What can we do now?

Dr. Mosca: Post-quantum algorithms are a critical first line of defense for our digital
systems. However, there is a non-negligible chance that they will be broken. So, for
long-lived information and for critical systems where the risk of cryptanalysis is too
high to just accept, it is imperative to have an additional layer of defense. QKD is a
remarkable solution for the key establishment problem, which is not susceptible to
unexpected mathematical attacks. It is already commercially available, and it’s a
technology that “ages well” — as quantum technology continues to advance, it
becomes easier and easier to overcome QKD’s current limitations.

evolutionQ pioneered the Quantum Delivery Network through its BasejumpQDN

software product, a very simple and elegant solution for integrating QKD-generated
keys into enterprise systems in a way that is scalable and vendor-independent.
Specifically, BasejumpQDN provides organizations with a framework to transition
from a low-cost testbed for simulated QKD to managing a complex multi-node
network of QKD. BasejumpQDN enables QKD to be deployed in a cost-effective
manner by expanding the number of nodes that can leverage QKD.

Cryptocurrency
Cryptography is the key underlying technology to the blockchain and contributes its
name to the eponymous cryptocurrency. There are of course numerous types of
cryptocurrencies. However, we will use Bitcoin as a proxy to discuss the potential
effects of quantum computing on the industry more broadly. To be clear, we are not
saying the introduction of quantum computing means Bitcoin or other
cryptocurrencies will be hacked. But we are pointing out the increased risk to the
cryptography of Bitcoin and other cryptocurrencies without the right technology
protocols put in place.

Bitcoin’s Cryptography
The Bitcoin network is secured by computers all around the world, referred to as
miners, that use SHA-256, a 256-bit cryptographic hash function whose name
stands for Secure Hash Algorithm-256. All Bitcoin transactions must be “confirmed”
by the network of miners before they are added to the blockchain, an immutable
ledger of all historical transactions including current ownership rights.

Bitcoin, for example, uses a different type of cryptography than the RSA encryption
that protects most of the internet. Transactions in the Bitcoin network are digitally
signed using the Elliptic Curve Digital Signature Algorithm (ECDSA), which is based
on Elliptic Curve Cryptography (ECC) that relies on the principle that it would be
intractable for a computer to deduce the private key from the public key.90
Fundamentally, breaking this cryptographic primitive would mean that securely
transferring ownership of one’s Bitcoin would no longer be possible.

Classical computers are not able to break RSA encryption. However, both RSA and
ECC can be broken by a quantum computer capable of running Shor’s algorithm.
Interestingly, it has been suggested that a quantum computer may be able to crack
ECC in less steps than RSA.91 Ultimately, using Shor’s algorithm, it would be
possible to calculate a private key and steal someone’s Bitcoin.

With various alternative architectures for quantum computers all competing against
one another, we can only very roughly estimate when quantum computers might
become able to break Bitcoin’s cryptographic signature scheme. Taking the 10-
minute (or 600-second) transaction time as a cut-off, the Centre for Quantum
Technologies at the University of Singapore estimated that that this could occur
anytime ranging from the late 2020s to the late 2030s.92

90 Pascal Urien, “Innovative Countermeasures to Defeat Cyber Attacks Against

Blockchain Wallets: A Crypto Terminal Use Case,” downloaded from arXiv, PDF, March
30, 2023;Certicom Research, “Certicom ECC Challenge,” PDF, last updated November
10, 2009.
91 John Proos and Christof Zalka, “Shor’s Discrete Logarithm Quantum Algorithm for

Elliptic Curves,” downloaded from arXiv, PDF, February 1, 2008.

92 Divesh Aggarwal et al., “Quantum Attacks on Bitcoin, and How to Protect Against

Them,” downloaded from arXiv, PDF, October 28, 2017.

Figure 29. Time to Break Signature Scheme for a Quantum Computer

The plot shows two estimates of the time in seconds required for a quantum computer to break the signature
scheme (red curves) as a function of time. It also gives more and less optimistic estimates (red striped lines).
Source: Divesh Aggarwal et al. (2017)

Cryptocurrency Mining
As well as Shor’s famed algorithm, there is another important quantum computing
algorithm that could create challenges for a large subset of cryptocurrencies —
specifically, those that reach consensus through Proof of Work (PoW). PoW is the
original consensus mechanism for cryptocurrencies and debuted in Bitcoin, which
still uses this methodology. PoW operates on the principle that transactions must be
“confirmed” by the network of miners before they are added to the blockchain.

Grover’s algorithm is a quantum computing algorithm that makes it far simpler to

search through unstructured data sets. Researchers have found that when applied
to the principle of mining cryptocurrencies, Grover’s algorithm allows for a quadratic
speed-up of finding the solution to a block and thus of winning that block in the
blockchain, resulting in an increase in the hash rate of whomever is using the
algorithm.93

Henceforth, in a post-quantum world, a Quantum-Equipped Actor (QEA) could gain

an unfair advantage by mining blocks faster using Grover’s algorithm. As the
blockchain is an immutable ledger of all historical transactions including current
ownership rights, one possibility is that a QEA could use this more efficient mining
technique to undertake a 51% attack. This is where a single person or group of
persons gains control of over 50% of a blockchain's hashing power, by secretly
creating their own chain in a PoW cryptocurrency until they have gotten into the
lead, and then publish their blocks to cause a reorganization of the public
ledger.This threat will apply to all cryptocurrencies that use PoW, but as Bitcoin is
understandably the most academically researched, we will examine what quantum
computing could mean for its PoW consensus.

93Iain Stewart et al., “Committing to Quantum Resistance: A Slow Defence for Bitcoin
Against a Fast Quantum Computing Attack,” Royal Society Open Science, Vol. 5, No.
180410, May 22, 2018.

The same paper from the University of Singapore compared the hash rate of the
total Bitcoin network to that of what a single quantum computer may be able to
produce (Figure 30). It suggested that prior to 2028 (in the more optimistic
estimate), there would not be any quantum computer with sufficiently many qubits to
implement Grover’s algorithm.

Figure 30. Hash Rate of Total Bitcoin Network vs. Single Quantum Computer

The plot shows two estimates of the hashing power (in hashes per second) of the Bitcoin network (blue striped
curves) vs. a single quantum computer (red striped curves) as a function of time. It also gives more and less
optimistic estimates and uncertainty regions (blue and orange areas). For comparison, the black dotted line
shows the hash rate of a single classical ASIC miner. The graph is plotted on a logarithmic y-axis, meaning that
the disparity between a quantum computer’s hash rate and that of the total Bitcoin network is likely to be several
orders of magnitude apart for many years to come.
Source: Divesh Aggarwal et al. (2017)

It is important to note that a quantum computer requires a highly specialized and

precisely controlled environment making it unlikely to be generally available for use
in crypto mining. Also, given the competitiveness of crypto mining, by the time
quantum computers become reliable enough to be used for mining, they are likely to
be adopted by the majority of miners and hence establish a network equilibrium.
This is something we have seen throughout the history of crypto mining — a
gradual hardware transition from humble laptop CPUs to more-specialized but
widely available and general-purpose gaming GPUs, and eventually to the current
highly specialized and purpose-built ASICs that dominate crypto mining today.

The Broader Cryptocurrency Ecosystem

While we have focused on Bitcoin as an example, as of 2023, there are nearly
9,000 cryptocurrencies, according to Statista.94 However, it would not just be each
cryptocurrency itself that would be at risk from a quantum attack, but everything
built on top of it.

94Raynor de Best, “Number of Cryptocurrencies Worldwide from 2013 to February

2023,” Statista, March 8, 2023.

A study conducted in 2021, when the total market cap of cryptocurrencies was a
little under $2 trillion, stated that a quantum attack on crypto could result in a 99.2%
collapse of value. Interestingly, while the immediate loss to cryptocurrency holders
in this scenario was estimated at around $1.9 trillion, the indirect losses to the
economy were projected to be a further $1.5 trillion.95 We anticipate these potential
indirect losses to the economy (relative to the immediate loss to cryptocurrency
holders) could grow significantly due to everything now being built on top of the
existing blockchain infrastructure, such as Decentralized Finance (DeFi).

However, DeFi is likely just the start. With the mainstream adoption of “layer 2”
solutions that build on the underlying “layer 1” cryptocurrencies, as well as the
increasing usage of decentralized apps (or dApps) built on cryptocurrency
platforms, the value at risk could grow much higher. Take, for instance, the recent
adoption of non-fungible tokens (NFTs), which are now finding uses well beyond just
the art world — from music streaming to property ownership.96 With the increasing
investment in Metaverse and Web3 projects that are intended to build a more
decentralized web using distributed ledger technologies (DLTs), breaking the
underlying cryptography would enable a QEA to claim ownership over any digital
assets on said DLTs, regardless of what specific DLT architecture is used.

Potential Courses of Action

As discussed in the “Cybersecurity” section, considerable thought is being put into
Post-Quantum Cryptography (PQC). It is likely that in the near future, long before a
sufficiently powerful QEA can exist, the Bitcoin community could agree on and
deploy a quantum-resistant signature scheme.

Such a change is unlikely to be controversial at the time and rather could be seen
as a part of necessary improvements in security protocols, which have been a part
of the natural progress and development of the industry since its inception. There is
the chance, however, of disagreement between various stakeholders about how to
implement changes to an existing blockchain. Neven’s Law shows us just how fast
improvements in quantum computers can occur (apparently almost overnight, in
fact). In the event of such a threat, necessitating agreement between various
decentralized stakeholders could increase the time it takes to adopt a quantum-
resistant form of cryptography.

Furthermore, several organizations are actively investigating quantum resistance in

cryptocurrencies in general, with many marketing themselves as developing a
quantum-resistant ledger in general or offering a post-quantum solution to Bitcoin.
However, at the point when quantum computers are powerful enough to be able to
break the cryptography of cryptocurrencies, they will likely be equally able to break
the encryption associated with more pressing matters, such as most of the internet
(including one’s banking details).

95 Arthur Herman, “Will Quantum Computers Burst the Bitcoin Boom?,” Forbes,

November 9, 2021.
96 Citi GPS, Money, Tokens, and Games: Blockchain's Next Billion Users and Trillions in

Value, March 2023.

With many countries “going dark” about their developments in quantum computing,
it may very well be blockchains run by countries that are first to pivot to quantum
resistance. For instance, central bank digital currencies (CBDCs), which many
countries are still exploring or planning to roll out, have fundamental advantages in
pivoting to quantum resistance. First, they are run by the central banks of countries,
which means they may be privy to knowledge from their respective governments,
earlier than industry players, about when quantum computers begin to pose a threat
to cryptocurrencies. Second, their centralization makes them less constrained by
the time it takes to necessitate agreement between stakeholders.

Understanding the Landscape and

How to Prepare
The use cases of quantum computing will grow over time, and it will impact
stakeholders in completely different ways during these stages. We recognize that
the industry impact of quantum computing may be almost as broad as the impact of
computing, and different types of stakeholders will thus need different information
and guidance. For this reason, we have divided this chapter into the three broad
stakeholder categories as follows:

1. Nation-States

2. Corporates

3. Market Participants

Within each of these stakeholder categories, we have evaluated the quantum

computing landscape and identified topics that we feel each type of stakeholder
should be aware of, as well as included a section on “How to Prepare.”

In our “Nation-States” section, we discuss:

 Government investment and how to measure the utility of such investment.

 Why the quantum computing education levels in the workforce matter.

 How quantum computing supply chains differ from those of classical computing.

 Ethical questions around the use of quantum computing.

In our “Corporates” section, we discuss:

 The total addressable market (TAM) forecasts for quantum computing.

 What value creation could look like within the quantum computing space.

 Why collaboration and the cloud are key to corporate adoption.

In our “Market Participants” section, we discuss:

 The quantum venture capital trends in terms of the capital raised and deal
activity.

 How the company and funding environment looks in terms of regional divides
and the most active market participants.

Nation-States
Overview
Countries are uniquely positioned to prepare for the rise of quantum computing as
an emerging technology. This is due to their ability to bring together experts from
otherwise distinct fields, as well as provide the long-term funding necessary for the
sector to grow to scale.

Each nation-state will have its own approach that takes into account both the state
of play of its current industries and its long-term strategic goals. Hence, in this
section, we focus on the areas that would be conducive to a national landscape in
which a nation-state could best capitalize on quantum computing.

The four broad categories we identified in our literature review are as follows:

 Government investment

 Workforce education

 Supply chains

 Ethics

We then tie together conclusions from our literature review in the section “How to
Prepare: A Holistic Quantum Computing Policy.”

Government Investment
All new technologies have an initial stage of development that requires considerable
investment before any commercial benefits can be achieved. This is particularly true
of quantum computers given the long development timeframes involved. Also,
unlike their classical counterparts, quantum computers offer no guarantees that
such investment will provide any return in the near term.

The Precedent for Nascent Technologies

While quantum computers are now moving out of laboratories and into commercial
applications, they are still in their infancy in terms of commercial development.
Private investors in quantum technology are likely to be motivated by potential
advances in short-term commercial applications. This makes it even more important
that nation-states provide long-term funding, particularly for the early development
activities that are necessary to keep the quantum computing industry competitive.97
This is because government funding is necessary to reach the stage where
quantum computers become practically useful for businesses, at which point private
investment should, in theory, sustain their development.

Government support is also not without precedent. We often think of the

technologies that we take for granted today as simply being the by-product of
competition between technology giants, but in many instances, they would not have
had the opportunity to flourish without initial government support.

97U.S. Government Accountability Office, “Considerations for Maintaining U.S.

Competitiveness in Quantum Computing, Synthetic Biology, and Other Potentially
Transformational Research Areas,” September 26, 2018.

For example, the internet would not have existed without the ARPANET network
built in the 1960s by the U.S. Defense Advanced Research Projects Agency
(DARPA) — in fact, it was the first network to transmit data in discrete chunks and
created the building blocks for the development of the Transmission Control
Protocol/Internet Protocol (TCP/IP) specification we still use today.98 Our current
global position system (GPS) can trace its origins to a constellation of just five
DARPA satellites, called “Transit,” used to locate U.S. Navy ships. Similarly, the
speech translation that forms the backbone of modern-day digital assistants was
developed with DARPA funding. Another example is European countries’ support of
the development of second-generation wireless networks based on Global System
for Mobile Communications (GSM) technology. More recently, one of the leading
providers of 5G technologies reportedly accessed $75 billion in different forms of
state support from the Chinese government.99

One of the key reasons that classical computing scaled in performance at the rate it
did over the past 60 to 70 years is that the products and services ecosystem that
used the latest classical computing technology grew at the same time. This allowed
the classical computing industry to, at the earlier stages, generate exponentially
more revenue and reinvest those profits into the research and development (R&D)
needed to build the next generation of classical computers. As the National
Academies of Sciences, Engineering, and Medicine points out in its report into the
progress and prospects of quantum computing, in order for a Moore’s-Law-type
continuous exponential growth in quantum computing to occur, there must be
exponential growth in investment, and there must be a similar cycle for quantum
computers as there was for classical computers — where smaller, cheaper, or more
easily accessible machines are commercially successful enough to grow investment
in the overall area.100 Absent of this dynamic, funding the development of each
country’s quantum computing ecosystem will become the responsibility of their
governments.

The Public Investment Landscape

Numerous countries have already begun implementing national strategies for

quantum technologies (which include quantum computing, communications, and
sensing). As shown in Figure 31 below, the variations in announced government
funding for quantum technology appear particularly stark. For instance, China
appears to have declared four times more funding than its nearest competitor.
Notably, The Quantum Insider, which compiled this data, have noted that the
funding estimate for China is based on a limited number of sources and subject to
significant dispute. At the same time, they also note that this number is unlikely to
be below $4 billion.

98 Duncan Graham Rowe, “Fifty Years of DARPA: Hits, Misses, and Ones to Watch,”
New Scientist, May 15, 2008.
99 Chuin-Wei Yap, “State Support Helped Fuel Huawei’s Global Rise,” Wall Street

Journal, December 25, 2019.

100 Emily Grumbling and Mark Horowitz, Quantum Computing: Progress and Prospects

(Washington, DC: The National Academies Press, 2019).

Figure 31. Overview of National Quantum Technology Initiatives Globally in 2023

Quantum Spend Commitments ($bn)

16 Market speculation up
to $20 billion -
14 unverified
12
10
8 Classified amount
expected to be
6 higher
4
2
0

Note: Represents "total commitments" and does not reflect current annual spend.
Source: The Quantum Insider

It is also important to note that these announced funding numbers are also over
different timescales. The above table shows both backward-looking numbers of
amounts already invested and forward-looking numbers of current intentions to
spend in the area. For example, it has been estimated that since 2014, the UK and
U.S. governments have on average been spending £100 million ($124 million) and
$500 million a year, respectively, on their quantum technology programs.101 Of
course, such estimates are likely to be incomplete because they will always exclude
the classified spending of departments such as the Ministry of Defence (MoD) in the
UK or Department of Defense (DoD) in the U.S. One of the risks of insufficient
funding by governments is the potential of losing talent in quantum computing to
other countries, discussed further below.

On the other hand, governments that engage with industry early will set themselves
up with a strong foundation in developing quantum computing technologies. In fact,
this is exactly what the UK government has been doing with its National Quantum
Technologies Programme (NQTP), which is made up of the eight partner
organizations listed below.

101 Currency conversions are as of April 17, 2023; Karina Robinson, “Companies Cannot

Afford to Be Left Behind in the Quantum Revolution,” Financial Times, July 12, 2022. For
more details see U.S. National Science & Technology Council, “National Quantum
Initiative Supplement to the President’s FY 2023 Budget,” PDF, January 2023.

Figure 32. Partner Organizations that Make up the UK NQTP

Source: Used with permission from EPSRC on behalf of the National Quantum Technologies Programme

Fundamentally, the NQTP describes its purpose as supporting ideas, innovation,

and investment to secure UK advantage and opportunities in the globally
competitive new quantum era.102 We reached out to the NQTP, and they explained
that “The NQTP was established in 2014 and represents £1 billion of public and
private investment. It is designed to bring together academia, industry, and
government to accelerate the translation of quantum technologies into the UK
marketplace and open up opportunities for British businesses to unlock new
capabilities that can make a real difference to our everyday lives.”

Figure 33. Achievements of the UK NQTP

Source: Used with permission from EPSRC on behalf of the National Quantum Technologies Programme

As well as supporting investment into research, innovation, skills, and technology,

the program also provides UK companies with important grant funding to help them
identify and develop uses and applications for quantum technologies. Figure 33
shows some of the UK NQTP’s achievements so far. More recently, the UK
government announced a National Quantum Strategy that is expected to more than
double its investment into quantum technologies, investing £2.5 billion over the next
10 years. However, government funding is only useful if it yields results —
something we discuss next.

102 UKNational Quantum Technologies Programme, “Transforming the World with

Quantum Technology,” accessed April 10, 2023.

Papers Published

Given the uncertainty of the direction of quantum computing technology and the
sheer variety of different approaches to it (even just in qubit technology, as we
touched on earlier in this report), it is difficult to compare different countries’ levels of
success objectively. One area we looked at as a potential early proxy of the success
of investment was the publication of research papers. While of course research
output will not necessarily correlate with successful government investment,
especially since much government funding is targeted towards industrial
commercialization, it does help in understanding the overall quantum computing
landscape.

Figure 34. Map of Top 10 Countries Producing Quantum Computing and Quantum Communication Papers

Source: U.S. Government Accountability Office

The U.S. Government Accountability Office (GAO) analyzed the trends in countries
producing research papers on quantum computing and quantum communications
over a 20-year period from 1996 to 2016. Their findings clearly showed that most
countries producing research in quantum computing also do so in quantum
communications. Furthermore, the results show that the Top 10 countries (as
measured by the number of research papers produced) in each of these fields are
broadly similar. As expected, there also seems to be a moderately strong correlation
between the amount of funding a country puts into quantum computing and the
number of research papers it produces.

Patent Applications

Another proxy of the success of government investment, is its ability to generate

intellectual property (IP) in the space, as measured by the number and quality of
patents it has filed. The patent application landscape provides a unique insight into
what is often the first stage of the commercialization lifecycle for any emerging
technology — near the point of new idea generation, and often before the stage of
receiving significant investment.

The ease with which different quantum technologies can be patented is incredibly
nuanced. Consequently, we looked solely at the quantum computing sector, to allow
for easier comparisons between countries. Leveraging the expertise of our in-house
data experts, Citi Global Data Insights (CGDI), we analyzed data provided by
QuantIP.

We found that there has been a substantial increase in global patent applications in
recent years, from just over 30 at the start of the last decade to an average of over
1,000 in the past two years. This has been mostly driven by a sustained growth in
the space since the middle of the last decade, with a 55% compound annual growth
rate (CAGR) from 2015 to 2022. This was around the time when the idea of the
commercialization of quantum computing was gathering momentum internationally.

Figure 35. Patent Applications Related to Quantum Computing (2010-22)

1600
1400
Patent Applications

1200
1000
800
600
400
200
0
2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022

Public Company Private Company Educational Institution

Source: Citi Global Data Insights, QuantIP

It was also not too long before various significant developments occurred in the
field, including claims of quantum supremacy. While it is not possible to draw direct
links to any specific developments, we feel that a healthy global patent environment
is critical to the successful commercialization of the technology. This is as it
provides a means for inventors to protect their ideas, ultimately giving them
confidence to invest in further research and development, and likely be more willing
to share those ideas with companies and governments.

Our analysis separated the different types of entities applying for these patents:
public companies, private companies, and educational institutions. We found that
while the number of patent applications from public companies grew 15x during this
period, patent applications from private companies grew 60x (and now make up
around 65% of all patent applications globally). This has most likely been facilitated
by the increasing levels of private investment into the space — something we
investigate further in our “Market Participants” section below.

In terms of the global distribution of patent applications by countries, it is of course

important to take into account how the quality of patents varies from one country to
another. For this reason, the CGDI team allocated patent quality weightings to each
country based on their assessment of patent filing robustness, market potential, and
citation potential. We then analyzed the global quality-weighted patent applications
by country for the period 2015-20. Our analysis found that the U.S. dominated new
idea generation, accounting for almost half (48%) of patent applications, followed by
Japan (17%), Canada (8%), China (7%), South Korea (3%), and the rest of the
world (17%).

Figure 36. Patent Applications Related to Quantum Computing by Country

100%

Quality-Weighted Patent Applications

90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
2015 2016 2017 2018 2019 2020 2021 2022
United States China Japan Canada South Korea Rest of World

Source: Citi Global Insights, QuantIP

What is most striking to see from the figure above is the pace of change in recent
years for certain countries. According to QuantIP, China’s share of global quality-
weighted patent applications grew from the aforementioned 7% to an average of
22% for the period 2021-22. This is even more impressive when considering that
the CGDI’s analysis attributed a quality-weighting to United States patents of 1.5x to
that of China for this period. During 2021-22, the U.S. still led globally at 45%, with
China in second place (22%), followed by Japan (8%), Canada (3%), South Korea
(2%) and the rest of the world (20%).

Given the global dominance of the U.S. and China in patent applications over the
past two years, we further investigated the distribution of patent applications by type
of entity. In 2021-22, the U.S. accounted for 65% of all the quantum computing
patent applications by public companies — most likely a reflection of the historical
strength and size of U.S. public markets. The proportion of patent applications by
private companies was generally split fairly equally between the U.S., China, and
the rest of the world. Interestingly, however, China accounted for 54% of patent
applications by educational institutions globally during this period. It is possible that
this could be an early sign of the high levels of government investment committed to
the space, or potentially a reflection of the traditional close links between
government and academia in the country.

Figure 37. Patent Applications Related to Quantum Computing by Type of Entity (2021-22)

Public Company 65% 3% 32%

Private Company 31% 35% 34%

Educational Institution 28% 54% 18%

0% 20% 40% 60% 80% 100%

United States China Rest of World

Source: Citi Global Data Insights, QuantIP

Ultimately, for both the U.S. and China, it seems there is notable strategic
significance being placed in new idea generation and securing intellectual property
in quantum computing.

Challenges for Government Investment

As with any nascent technology, high levels of government investment do not

guarantee success, even if that country is able to successfully publish academic
papers or apply for patents. Quantum computing is a very specific technology that
requires specialist knowledge. Hence, one of the key questions facing nation-states
regarding building their own quantum computing ecosystems is not just how much
funding they can provide, but whether there exists a sufficient enterprise base to
absorb it.

The UK, for instance, has a spin-off culture whereby academic departments with
expertise in the quantum computing field set up their own enterprises. It is dynamics
such as these that will play a key role in encouraging investment from the industry
in the long run, and as discussed earlier, are necessary for the long-term
development of the quantum computing industry beyond its incubation period.
However, in order for technology development to continue in the meantime (before
quantum computers offer a commercial advantage to businesses), governments
must face the challenge of ensuring businesses have confidence in the technology’s
commercial potential.

We spoke to Roger McKinlay, Head of the Quantum Technologies Challenge for UK

Research and Innovation (UKRI), about some of the difficulties governments face
today when investing in quantum computing.

Expert Interview with Roger McKinlay, Head of the

Quantum Technologies Challenge for UK Research and
Innovation
Q: Given the high levels of private investment in quantum, why should
governments continue to invest?

Roger: Governments need to act as co-investors, their investments being the

vehicle by which they engage with industry and have influence over how the
technology develops. It is not just about risk. It is now clear that private investors
are prepared to take the risk. The issue is more about strategic co-investment —
the public and private investment communities working together.

Q: What determines the level of funding that can be absorbed in a particular

country?

Roger: There are many factors. Fundamentally, the growth of the sector will depend
on the creation of new companies (spinouts or start-ups) and the rate at which
existing companies see quantum technologies as being part of their future business.
It does not all have to be “pulled” by end-use. Individual businesses just need a
Roger McKinlay
“customer” who may or may not be at the end of the supply chain. Some countries
Head of the Quantum Technologies
may excel at a component level, and others in quantum computing applications.
Challenge, UK Research and Innovation
Q: Where does the international dimension come in?

Roger: There are two angles on this. Firstly, unless states want to grow their own
start-to-finish supply chains, international trade will be essential. As with any
expensive technology, companies will need access to global markets to make the
case for the investment in developing new products and services. Secondly, the
capabilities quantum technologies bring will be considered “sovereign” by many
states. The issues will be what to “own,” where to “partner,” and what to “acquire.”
We will see groups of “trusted traders” form.

Q: To what extent will the availability of skills determine the growth of the
sector?

Roger: In general, companies are good at competing for the talent they need to
make their businesses work. They will recruit, train, and develop people as
necessary. They will compete with other companies, not just for business but for the
skills to deliver it. Governments can help by removing restrictions and by making
sure that this skills and talent market really can function as a market. In the long
term, the national education system needs to be able to deliver people who are
capable of acquiring the necessary skills if the growth is to be sustained. STEM
subjects will be key. At the moment, we tend to look at public investment as a
measure of how seriously countries view quantum. In the longer run, the skills issue
will dominate how and where the industry scales up.

Workforce Education
As with any emerging technology, in order for a nation-state to leverage its benefits
effectively, it needs not only to invest in it financially, but invest in educating its
workforce to ensure there is a sufficient supply of talent to drive further
technological advancements. The challenge with quantum computing is that there
have been so many significant developments in recent years, and the industry has
grown so fast, that the supply of talent is becoming a bottleneck. Many believe this
talent bottleneck risks becoming a significant barrier to the industry’s growth in the
long run.

Understanding the Talent Shortage

Working in quantum computing is quite different to other industries — for instance,
many in the field have come directly from academia, making the talent pipeline quite
different than in established disciplines like data science or cybersecurity. This is
because the quantum computing industry is not yet mature enough that most
individuals cannot do without at least a basic understanding of quantum mechanics.
There is a close link between quantum hardware and software, which are more
intricately connected than their classical counterparts. Consequently, experts need
to have a sufficient understanding of the underlying physics that give rise to a
qubit’s special properties in order to be able to design quantum hardware and
quantum software accordingly.

What this means is that, up until now, progress in the quantum computing sector
has been mostly achieved by PhD-level scientists. This is something that continues
to be reflected in today’s job’s market.103 One recent study analyzed job posts on all
quantum technologies (i.e., computing, communications, and sensing), of which
42% were focused in the quantum computing sector specifically. Investigating over
750 vacancies globally, researchers found that on average 61% were being
advertised as requiring a PhD — although with some global regional variations with
U.S./Canada (at 52%) and Europe/Other (at 67%).

Figure 38. Degree Requirements of Jobs in Quantum Technologies Figure 39. Professional Experience Requirements of Jobs in Quantum
Technologies
None 100%
5% 90%
80%
70%
60%
Bachelor's 50%
27%
40%
30%
PhD
20%
61%
10%
Master's 0%
7% PhD Master's Bachelor's None
1-3 Years 3-5 Years > 5 Years

Source: Kaur and Venegas-Gomez (2022) Source: Kaur and Venegas-Gomez (2022)

103Maninder Kaur and Araceli Venegas-Gomez, “Defining the Quantum Workforce

Landscape: A Review of Global Quantum Education Initiatives,” Optical Engineering,
Vol. 61, No. 8, May 19, 2022.

The study also highlighted how significant holding a PhD was in terms of mitigating
the additional years of industry experience required to fill roles. While only 20% of
roles that required a PhD asked for more than 3 years of experience, this increased
drastically to 50% and 70% for those only holding up to a Master’s or Bachelor’s
degree, respectively — something that is likely due to the research skills and
industry-specific knowledge gained through doing a relevant PhD. While a small 5%
of the overall quantum-related roles did not advertise any form of higher education,
almost 90% of these roles asked for more than the above mentioned 3 years of
experience.

However, this requirement for a PhD is not universally distributed across all types of
jobs in the industry. While, of course, some roles such as a theoretical or
experimental physicist, would necessitate such a background, as Figure 40 below
shows, there a number of other roles, such as maintenance technicians,
measurement engineers or circuit designers, that typically only require a Bachelor’s
or Master’s degree.

Figure 40. Distribution of Degree Requirements for Different Jobs in Quantum Technologies

Source: Hughes et al. (2021) 104

104Ciaran Hughes et al., “Assessing the Needs of the Quantum Industry,” downloaded
from arXiv, PDF, August 25, 2021.

Ultimately, this has led to the very well-known problem of the quantum talent
shortage, whereby governments and companies in the quantum technologies space
have been struggling to find individuals with the appropriate level of quantum-
relevant skills. One report claimed that the number of quantum physicists and
engineers was estimated at only a few thousand worldwide — and there is certainly
no guarantee that they would all end up in any of the quantum technology
industries.105 Many feel this talent shortage risks becoming a significant hurdle to
the growth of quantum computing in the long run, with “a lack of talent” ranking third
in a 2021 survey among all the possible barriers to the development of the
technology.106 In fact, it has been reported that the skills shortage in quantum
computing could even harm the UK economy unless universities begin to recruit
more students.107 However, thankfully, there appears to be a changing tide in the
educational requirements to join the quantum computing workforce.

An Evolving Educational Landscape

There seems to be an overwhelming consensus from all the experts we spoke to

that, for the quantum computing industry to scale at speed over the coming years, a
more holistic approach to hiring needs to be taken. As Professor Sougato Bose put
it when we interviewed him (see earlier in this report): “Quantum mechanics itself is
typically taught, at least in the form needed for quantum computation, to
undergraduates taking physics as a major, only in the last or third year. It has not
spread out to the curriculum of other disciplines, as well as to earlier levels of
physics curriculum in terms of the very fundamentals such as qubits and gates. I
think this will be a change that will be necessary to have a larger volume of
educated workforce in this area…”

This goes someway to further explaining demand for PhDs in the quantum
computing industry — as a PhD has often been the first real point at which a
student could begin to specialize in quantum technologies and gain the industry-
specific knowledge needed by employers. Fortunately, it seems that universities
around the world are recognizing this, and increasingly offering Master’s degrees
specifically in quantum technologies.

105 Laura G. Converso, “How to Build a Quantum Computing Workforce,” Accenture,

November 9, 2020.
106 Brian Lenehan, “Quantum Talent — Shortages and Tactics,” Quantum Strategy

Institute, December 3, 2021.

107 Mark Piesing, “’How Can We Compete With Google?’ The Battle to Train Quantum

Coders,” The Guardian, January 15, 2020.

Figure 41. Number of Master’s Programs Focused on Quantum Technologies Across The World

Source: Kaur and Venegas-Gomez (2022)

It is important to note that these are still rare, with most countries only having one to
two relevant Master’s degrees in their entire graduate-level educational system, but
nonetheless is a step in the right direction to lowering the academic barriers to
entering the industry. The United States and the United Kingdom seem to again be
leading in this respect with 11 and four Master’s courses respectively, as of the time
of publication. However, even with the increasing prevalence of Master’s programs,
there remains numerous challenges. There has been criticism of some curriculums
merely packing courses relevant to quantum computing together without giving
serious thought on the integration of these separate subjects, and ultimately
requiring companies to do extra training. This may be contributing to why employers
are requesting 3+ years of industry experience in 50% of Master’s-level job
openings, as mentioned above.

There have also been increasing calls for more engineering talent to be generated
to supply the individuals who will be able to turn early and low-TRL (Technology
Readiness Level) quantum experiments into commercial quantum products and
services. In 2019, a symposium of 50 quantum experts noted that the knowledge
required by existing quantum-related academic curriculums was not practical
enough for graduates to apply in the industry.108 These concerns are requiring
educational providers to increasingly have to think about more practical quantum
engineering skills, and as the industry becomes more mainstream, we are more
likely to see dedicated undergraduate programs in this area. In fact, there is already
work underway in designing courses from the ground up to meet the practical needs
of working in the quantum technologies industry. These include educational
opportunities offering more hands-on training with quantum hardware, including
those relevant to different modalities of optics, ions, nanofabrication, and different
tools required in the industry.109

108 CyberTalk,
“Closing The Quantum Computing Talent Gap,” August 4, 2021.
109Abraham Asfaw et al., “Building a Quantum Engineering Undergraduate Program,”
IEEE Transactions on Education, Vol. 65, No. 2, May 2022.

Nonetheless, this is still a far higher barrier to entry than for many other industries,
and there is much work to be done to make the prospect of a career in quantum
computing more accessible for much of the workforce. Nation-states are
increasingly realizing this and incorporating the matter of workforce education into
their national strategies. The U.S. has already begun this process and suggested
ways to develop a quantum-ready workforce. This includes actions that intend to
introduce broader audiences to quantum technologies through public outreach and
education materials, address quantum-specific gaps in professional education and
training opportunities, and make careers in quantum technologies more accessible
and equitable.110

All this, however, will take time to implement. Even without the need for a PhD
(which, as discussed above, most jobs still require), given it typically takes 3-5 years
to train a new student and meet the employment expectations to work in the field,
such initiatives are unlikely to provide a solution to the quantum computing skills
shortage in the near term.

Training Those with Adjacent Skills

To meet the near-term demand for talent in quantum technologies, it will be

increasingly necessary to train those with adjacent skills. Workers will likely be
generalists with backgrounds in disciplines such as computer science or
engineering, rather those educated specifically in quantum technologies.111 One
study found that there could be as many as 350,000 Master’s-level graduates
globally in fields that are relevant to quantum technologies, with over 100,000 in the
European Union alone.112

Figure 42. Number of Graduates in Fields Relevant to Quantum Technologies

160,000 350

140,000
Graduates in Quantum Technology-

300

Density Per Million Inhabitants

120,000
250
Relevant Fields

100,000
200
80,000
150
60,000
100
40,000

20,000 50

0 0
European India China United Russia United
Union States Kingdom

Number of graduates of master’s level or equivalent in 2019 in biochemistry, chemistry, electronics and chemical
engineering, information and communications technology, mathematics and statistics, and physics. High-level
estimates are represented for China. The actual U.S. talent pool may be larger, as bachelor’s programs are
longer and master’s programs are less common.
Source: McKinsey Digital, Citi GPS

110 U.S. National Science & Technology Council, “Quantum Information Science and
Technology Workforce Development National Strategic Plan,” PDF, February 2022.
111 Chuck Leddy, “Q&A: The Talent Shortage in Quantum Computing,” MIT News,

January 23, 2019.

112 McKinsey Digital, “Quantum Technology Sees Record Investments, Progress on

Talent Gap,” April 24, 2023.

Such graduates would still need training to adapt their skills to make useful
contributions to quantum engineering-related work. Numerous one-year
supplementary curriculums are being considered to help with switches to quantum-
related roles, but also several short courses are beginning to emerge to allow the
transfer of skills to the quantum computing industry.

Multiple public and private entities are engaged in these efforts. For instance, MIT
offers a number of online courses, ranging from those providing an introductory
understanding of quantum computing fundamentals, to more specialist courses
looking at different types of quantum algorithms. There are also efforts by various
organizations, such as The Quantum Strategy Institute, to educate future graduates
on the requirements of quantum engineering roles.113 Other examples include Q-
CTRL’s Black Opal learning tool which offers a training portal that individuals can
use to transfer their skills to quantum computing, and IBM’s quantum developer
certification program.114

In addition, meaningful internship and training programs that last at least 3-4
months can also help expand quantum computing candidate pools. IBM has been
running a quantum computing internship with an annual intake of well over 100
people for several years.115 Furthermore, hiring agencies are already building their
own quantum computing candidate pools.

Of course, it is important to note that not all types of jobs in quantum computing
need quantum-specific skills. In all likelihood, as the industry matures into a phase
driven less by research and more by product and service delivery, the proportion of
jobs in the industry that require more commercial and industrial skills, such as
project management or marketing, is likely to grow. Some in the industry have
suggested that roles that do not need quantum-specific skills (e.g., general software
developers, non-technical roles such as product sales, etc.) are already becoming
as in-demand as those requiring quantum-specific skills (e.g., experimental
physicists, quantum algorithm developers, etc.).

Fundamentally, the upskilling and re-skilling of people with adjacent skills will play a
critical role in meeting the demands of the quantum computing industry. However,
as the Director of the National Science Foundation (NSF) described in a recent
quantum technologies conference, there may not even be enough trainers in the
quantum computing industry to upskill or re-skill a quantum-relevant workforce. One
report suggests that less than 50% of quantum computing jobs will be filled by
2025.116 As such, the talent shortage is likely to persist for some time, with wage
inflation and competition for talent seeming likely — yet another reason why nation-
states wishing to adopt the technology need to plan ahead.

113 Amrita Manzari, “Roadmap to Quantum Engineering,” Quantum Strategy Institute,

October 26, 2021.
114 Jeffrey Burt, “Q-TRL Black Opal: Quantum Learning for the Masses, QControl’s Black

Opal Learning Tool” The New Stack, December 3, 2021; Abe Asfaw, Kallie Ferguson,
and James Weaver, “IBM Offers Quantum Industry’s First Developer Certification,” IBM,
March 29, 2021.
115 IBM, “IBM Quantum Internship Applications for Summer 2022 Are Still Open,”

September 7, 2021.
116 Niko Mohr et al., “Five Lessons from AI on Closing Quantum’s Talent Gap — Before

It’s Too Late”, McKinsey Digital, December 1, 2022.

National Security Concerns

Finally, one particularly challenging issue for nation-states is that much of the talent
in the field comes from foreign countries. A report from the White House
Subcommittee on Economic and Security Implications of Quantum Science (ESIX)
emphasized the indispensable role foreign talent and international companies play
in the U.S. quantum ecosystem.117 However, despite this, due to the fact that
quantum computing has been recognized as an area of strategic significance to
many countries, even if these foreign experts agree to join a company in the U.S.,
they must go through immigration reviews conducted by the government to be
officially onboarded.

The ESIX report underscored that the U.S.’s domestic quantum computing talent
supply fell short of domestic workforce demands by a large margin. With the
competition to deliver large-scale quantum computers getting increasingly fierce,
this shortfall of talent is only likely to increase in the near term. Issues like this
present both a challenge and an opportunity for nation-states, as governments can
help foster the needed collaboration between academia and industry.

We spoke to Dr. Stefano Gogioso, a leading computer scientist and researcher in

quantum computation at Oxford University, about some of the challenges in
educating a workforce in quantum computing.

117 Nick Flaherty, “U.S. Summit On Quantum Industry Highlights Skills Shortage,” EE
News Europe, October 11, 2021; National Science & Technology Council, “The Role of
International Talent in Quantum Information Science,” PDF, October 2021.

Expert Interview with Computer Scientist and Researcher,

Dr. Stefano Gogioso, at Oxford University
Q: What are the biggest challenges when it comes to educating a workforce in
quantum computing?

Dr. Gogioso: Quantum computers are exotic machines. They are fundamentally
different from the computers we use today and scaling them beyond small
prototypes has proven challenging. Such development efforts often dominate the
quantum news cycle, but they are not actually a concern for the workforce at large:
Figuring out which hardware architectures scale best, among the ten or so major
ones currently in development, is a task for highly specialized physicists and
engineers.

When it comes to educating and training professionals, the biggest challenge is that
the exotic nature of these machines extends beyond their hardware, to the
problems they can solve and the way in which they solve them.

Quantum computing is going to be a truly differentiating factor for businesses and

Dr. Stefano Gogioso
nation-states only in the presence of custom applications: In any such adoption
Computer Scientist and Researcher in
scenario, professionals are inevitably faced with an entirely different way of
Quantum Computation, Oxford University
computing and solving problems. Unfortunately, the language used by quantum
specialists is decades old and heavily reliant on advanced mathematics, making it
inaccessible to the wider public: As a first step, we need to rephrase the same
concepts in new and relatable ways. Then, we need to develop programming tools
that use these ideas to make quantum computing accessible to professionals from a
broad variety of backgrounds. The only way to educate a workforce at scale is to
give them the means to truly understand how quantum computing works, to easily
experiment with it, and to quickly prototype new applications.

Q: What level of upskilling is expected for professionals wishing to join the

quantum computing workforce?

Dr. Gogioso: The simplest scenarios for quantum adoption involve the solution of
problems from a pre-defined class, using a pre-packaged quantum computing
service (D-Wave's quantum annealing is an example). Here, professionals are only
required to moderately upskill, learning to recognize suitable problems (for D-Wave,
constraint satisfaction and combinatorial optimization) and being trained to
formulate such problems for submission to the service (for D-Wave, this happens in
the form of quadratic unconstrained binary optimization — QUBO). Existing domain-
specific knowledge remains the primary driver for the origination of interesting
problems, as well as the creation of new value using solutions obtained from the
quantum computers.

More sophisticated adoption scenarios instead require the formulation of custom

applications for quantum computers: These could range from novel uses of
established quantum algorithms to the development of brand-new ones through a
continuum of proprietary modifications. For professionals wishing to develop such
applications, the upskilling is significant but progressive. In the chemical,
biomedical, and materials sectors, for example, one could start by learning about
relevant quantum simulations and how to tailor them for execution on quantum
computers.

For machine learning and artificial intelligence professionals, one could instead
learn about quantum neural networks, where they are expected to outperform their
classical counterparts, and the idiosyncrasies of their usage. Specific training is first
required to write effective quantum programs, then to make efficient use of
expensive computing resources, and finally to squeeze the most out of available
quantum hardware. Experimentation and the drive to extract value will, in time, lead
to the customization and optimization of existing quantum techniques, and then to
the development of new ones, informed by domain-specific knowledge and
business needs.

Supply Chains
Supply chain challenges have become familiar to nearly all large corporations over
the past two years. In particular, the silicon chip shortage has shown just how fragile
the global supply chain is, and how unexpected events can disrupt what many
would consider one of the world’s most heavily invested in and broadly
interconnected industries — classical computing.

What Makes the Quantum Computing Supply Chain Different

Like that of classical computing, quantum computing’s supply chain is global, but
even more highly specialized and complex. As highlighted in a July 2016 report by
the National Science and Technology Council titled Advancing Quantum Information
Science, the additional needs of quantum computers over classical computers
involve the development of novel qubit technologies, as well as other technologies
needed to control a quantum computer. All these various quantum technologies
could result in supply chain bottlenecks of strategic importance to nation-states.
Complicating this further is the fact that different types of quantum computers with
different qubit technologies are currently being pursued.

On the one hand, the large variety of qubit technologies could make it more difficult
for manufacturers to achieve the economies of scale reached by the classical
computing industry, thereby inhibiting manufacturers’ ability to lower prices. On the
other hand, this same breadth of approaches to building quantum computers could
mitigate the increase in prices that could arise from a global quantum computing
“gold rush.” Right now, it is too early to say if there will even be a winning qubit
technology that achieves adoption across the world (in the same way that the silicon
chip became predominant in the manufacturing of classical computers), or to predict
the impact this may have on a still-developing quantum computing supply chain.

Novel qubit technologies are just one part of the overall quantum computing supply
chain, which countries are already beginning to evaluate more holistically. The U.S.
Government Accountability Office (GAO), for instance, produced the chart shown in
Figure 43 highlighting some of the components needed to make a quantum
computer that are supplied from outside of the U.S.

Figure 43. Select Examples of Quantum Technology Component Parts Around the World

Source: U.S. Government Accountability Office

As Figure 43 shows, in addition to rare earth materials from China and the
semiconductor technologies from Taiwan that already present significant
bottlenecks for the classical computing supply chain, components specific to the
quantum computing supply chain are produced all over the world. One example
identified above is the manufacturing of dilution refrigerators in Finland (although
notably, since the production of this map, numerous other countries have begun
looking into building their own).

The Need for Extreme Cooling

Originally proposed by Heinz London in the 1950s, dilution refrigerators are able to
cool their contents to near “absolute zero.”118 Absolute zero is the coldest possible
temperature of any material; at that point, the material contains in effect zero energy
and corresponds to around −273.15 °C on the Celsius temperature scale. The
reason reaching this temperature is important for the manufacturing of quantum
computers is because, as discussed earlier in the report, qubits are susceptible to
noise, which comes in the form of ambient thermal energy. Thus, quantum
computing technologies, such as superconducting qubits, need to be supercooled to
these extremely low temperatures to prevent decoherence.

118 GrahamBatey and Gustav Teleberg, Principles of Dilution Refrigeration: A Brief

Technology Guide, Oxford Instruments NanoScience, PDF, September 2015.

When discussing temperatures this low, most people refer to the Kelvin (K) scale —
an absolute temperature scale obtained by shifting the Celsius scale by −273.15 °C
so that absolute zero coincides with 0 K.119 In fact, the temperatures needed by
quantum computers are sometimes measured in milli-Kelvin, or mK, only a few
thousands of a degree above absolute zero. For context, the temperature of outer
space is often quoted as being around 2.7 K.120 Ultimately, this need for extreme
cooling presents a potential additional bottleneck for the already complex quantum
computing supply chain.

Figure 44. Quantum Computing Dilution Refrigerator

Source: Shutterstock

We spoke to Dr. Anthony J. Yu, Vice President of Silicon Photonics Product

Management at GlobalFoundries, a semiconductor foundry that is collaborating with
industry leaders to produce photonic-based quantum computers, about some of the
unique characteristics of the quantum computing supply chain.121

119 Encyclopedia Britannica, “Temperature: Physics,” last updated December 1, 2022.

120 Philip Ball, “Space: How Cold Does It Get When We Leave Earth?” BBC, September
19, 2013.
121 Prableen Bajpai, “An Overview of the Top 5 Semiconductor Foundry Companies,”

Nasdaq, October 1, 2021; GlobalFoundries, “GlobalFoundries Announces Next

Generation in Silicon Photonics Solutions and Collaborates With Industry Leaders to
Advance a New Era of More in the Data Center,” March 7, 2022

Expert Interview with Dr. Anthony J. Yu, Vice President of

Silicon Photonics Product Management at
GlobalFoundries
Q: What are the key differences in the quantum computing supply chain
compared to the classical computing supply chain?

Dr. Yu: The key difference in quantum-based compute is that most, if not all,
implementations involve dilution cooling to 4 degrees Kelvin (K) or less, require 3D
semiconductor packaging compatible with large form factors and very low
temperatures (<4K to reduce phonon noise), and tend to use non-standard
semiconductor materials (e.g., yttrium, magnesium, niobium nitride, barium tantalum
oxides, and nitrides) for either single ion sources or single photon detection.

Semiconductor lasers have been used in strong attenuation mode as single photon
sources and should be included in the supply chain (III-V lasers). Novel/rare
materials are covered by the GAO map — advanced low temperature 3D
semiconductor packaging is not. The ideal form factor for a quantum system would
be to fit within a 2U or 4U datacenter rack — which drives extremely miniaturized
Dr. Anthony J. Yu
cooling systems (first solutions will be dedicated cabinets and not 2Us).
Vice President of Silicon Photonics Product
Q: What do you think are the key challenges to the quantum computing
Management, GlobalFoundries
supply chain?

Dr. Yu: This is, by definition, a low temperature application — every quantum
system that we are aware of requires cooling of the source and the detection to
somewhere between 4 millikelvins (mK) and 4K — the strongest supply chain
disruption would be access to cooling solutions.

A business challenge with developing the quantum supply chain is volume — these
solutions are in their embryonic stage at the present time, they will need volume
drivers to move from new product introduction (NPI) to full production. To facilitate
this transition, due to the uniqueness of the quantum solutions, capital investment
will be needed ahead of time — with strategic vision — or the volume ramp and the
promise of quantum compute will be delayed. In this sense it is no different than any
other 300mm investment opportunity — aside from the very early product maturity.

Q: What do you think stakeholders should be aware of about the quantum

computing supply chain?

Dr. Yu: The packaging and 3D wired or wireless interconnect to quantum compute
elements is fundamentally different than mainstream compute. This requires
dedicated investments (usually IDMs) to implement the specialized systems. There
is little to no economy of scale due to the uniqueness of the approaches to quantum
compute — targeted investment ahead of the need is required.

Ethics
With great power comes great responsibility. Quantum computers have the potential
to reshape our world by enabling breakthroughs in medicine, material science,
finance, and other industries. This implies, however, that they could unleash equally
great terror if used improperly. With breakthroughs in medicine design come
potential breakthroughs in the design of chemical weapons, for example.

The Need for Ethics

Such risks underscore the importance of players in the quantum computing field to
proactively consider the ethical dimensions of building such machines. In particular,
as nation-states race to be the first to build a fault-tolerant quantum computer,
having a set of ethics, rules, and guidelines may be critical to ensuring that quantum
computing remains a boon to, and not a burden on, humanity.

The U.S. government has been an early champion of ethical considerations in the
field of quantum computing. Alongside dedicated R&D funding, multiple branches of
government have taken action to ensure that the quantum programs and grants
incorporate ethical considerations in their structure.

Figure 45. Ethical and Legal Considerations

Source: Shutterstock

Integrating Ethics into Legislation

One year after the 2019 National Defense Reauthorization Act (NDAA) authorized
the Department of Defense (DoD) to create a Quantum Information Science (QIS)
research, development, and deployment program, Congress added a mandate for
ethical considerations in the NDAA.122 This provision requires the DoD to develop a
plan for the “development of ethical guidelines for the use of quantum information
science technology.”123

122 U.S.
Congress, “H.R. 5515,” PDF, accessed March 8, 2023.
123 U.S.
Congress, “National Defense Authorization Act for Fiscal Year 2020,” PDF,
accessed March 8, 2023; Congressional Research Service, Defense Primer: Quantum
Technology, updated November 15, 2022.

Ethical considerations have also been embedded in new government programs from
their inception, for example in the recently signed CHIPS and Science Act. This
legislation mandates that the Director of the National Science Foundation (NSF) revise
proposal instructions — within 24 months — to require that ethical considerations are
included in applicable proposals for funding, and to instruct that these ethical
considerations be factored into award decisions.124 This provision begins with the notion
that emerging areas of research present potential ethical concerns, and other sections
of this law identify quantum technology as an key technology focus area for the
authorized programs. Therefore, it is likely that the NSF Director will apply the
requirement for ethical considerations to be included in quantum computing grant
proposals.125

The CHIPS and Science Act also includes ethics in its creation of the new NSF
Directorate for Technology, Innovation, and Partnerships (TIP), the $81 billion
program designed to spur “translational and use-inspired” research and
development in the ten listed key technology focus areas, which include quantum
information science and technology.126 This provision mandates the Director to
ensure that ethical considerations are “taken into account in the priorities and
activities of the [TIP] Directorate, including in… the awards-making process and
throughout all stages of supported projects.”127

These two examples of recent statutory and regulatory action on quantum

computing ethics provide a sense of current quantum computing ethical priorities in
the U.S. government, and hint at what we can expect in the near future from other
governments. First, the above authorized programs need to be implemented in full,
which will likely involve consultation with non-government stakeholders and could
involve notice and comment rulemaking. Other federal agencies not described in
detail here, such as the National Institutes of Standards and Technology (NIST) may
join the NSF and DoD in establishing ethical standards or guidelines for quantum
computing. Finally, the 2018 National Quantum Initiative (NQI) Act is due for
reauthorization this year, and the inclusion of quantum ethics in the NDAA as well
as the CHIPS and Science Act suggests that Congress may look to include ethical
considerations in the NQI reauthorization.

In addition, many within the field of quantum computing itself have begun to consider the
ethics of building such machines. Fundamentally, the earlier that nation-states begin
legislatively addressing the issue of ethics in quantum computing, the better we can
prevent ethical issues arising from the technology’s use in the future.

We spoke to Nick Farina, CEO of EeroQ and quantum ethics advocate, about how
the issue of ethics in quantum computing is being addressed.

124 U.S. Congress, “H.R. 4346,” PDF, accessed March 8, 2023.

125 Ibid.

126 Ibid.
127 Ibid.

Expert Interview with Nick Farina, CEO of EeroQ

Q: Why should we consider ethical guidelines for quantum computing,
despite the relative immaturity of the technology?

Nick: Quantum computing will open entirely new frontiers of processing power,
which we are only now beginning to understand. To create ethical guidelines for the
use of this technology, we must first focus on understanding all of the potential
applications (both positive and negative) of quantum computing. This is a nontrivial
task, involves both quantum researchers and end-users of the technology, and will
always be a work in progress as new applications are discovered. Second, once we
understand the potential applications, we then need to think carefully about
developing ethical frameworks around them. This is a process that will take multiple
years and involve coordinating agreement among myriad stakeholders, both
corporations and governments. Thus, the process should begin now and in parallel
with quantum computer development.

Nick Farina
Q: How can companies engage in federal policy around quantum computing
CEO, EeroQ
ethics and regulation?

Nick: It is our view that quantum computing needs a multidisciplinary approach to

regulation, as it will impact such a broad swath of society. Regulation at the federal
level is inevitable and has already begun with pending export controls and the
requirement for the Department of Defense to develop ethical guidelines in the 2020
National Defense Authorization Act. For companies looking to make the most of
quantum computers while ensuring their usage minimizes harm, this formative time
of legislative activity provides a chance to have an outsized voice in the dialogue.
Ideally, regulation will avoid overly onerous restrictions on commercial use, while
still preventing either clear misuse of the technology by bad actors or accidental
abuse of the technology by those with good intentions but no regulatory
frameworks.

Q: How can novel types of quantum computing designs (such as neutral

atoms, photonics, or electrons on helium) change the expected timeline of
quantum computing’s impact and related ethical concerns?

Nick: If we assume the progress of quantum computing power to be linear, then

one might expect it will be 10 or more years before we have applications requiring
ethical considerations. However, we have many points of evidence that the
technology will develop at a faster, more nonlinear pace than expected. One
particular “X factor” is the rise and funding of new types of quantum computers that
offer the potential to “leapfrog” the technologies currently furthest ahead. For
example, the first two-qubit gate of an ion trap was in 1995, and ion traps are still in
the range of dozens of qubits. While this has not been physically realized yet, there
are multiple credible and well-funded academic and industry efforts that may be
able to scale from the order of two to 10,000 qubits much faster than today’s most
well-known technologies. If one of these technologies begins to work, we will face
ethical challenges much sooner than expected.

How to Prepare: A Holistic Quantum Computing Policy

In conclusion, quantum computing offers nation-states an opportunity to capitalize
on the potential next leap in computing. While the best-positioned countries are the
ones investing in the infrastructure and educating their workforces now, there is still
considerable opportunity left given the early stage the industry is in. What is clear,
however, is that any nation-state wishing to build its quantum computing industry
needs to establish a holistic policy that supports both the underlying technology
infrastructure and the overall quantum computing ecosystem.

As discussed earlier, government investment is often key to getting a nascent

technology off the ground. This is especially true of quantum computing given the
technology’s specialized nature and the fact that quantum advantage has yet to
occur. As the technology develops further, the talent shortage is likely to worsen in
the short term — similar to how many countries have experienced a shortage of
software engineering talent in recent years. However, unlike in that shortage,
gaining the skills to become a quantum-computational engineer or scientist is likely
to take many years. This means that nation-states need to start thinking ahead, and
in order to meet future demand, begin to design curriculums at much earlier
academic stages than is currently the case. This will require collaboration between
different governmental agencies, academia, and ideally, industry.

Of course, bottlenecks in the quantum supply chain are a big risk, as components
are sourced from all over the globe (see Figure 43). Hence, strategic initiatives
between countries in this decade may be crucial to a nation-state’s ability to
capitalize on quantum computing in the decades ahead. Finally, forward-thinking
nation-states should be considering quantum computing’s attendant ethical
concerns before we reach the era of quantum advantage.

The 2021 Citi GPS report Holistic Digital Policy: Nation States Must Lead in Building
Equitable Human-Centric Digital Economies analyzed the digital transformations of
11 countries and effectively created a toolkit that governments around the world
could use in designing their own national digital policies. It then looked into the
approaches individual countries could take around digital policy based on their
unique strengths and provided a framework for implementing a holistic policy. Given
the strategic significance of quantum computing to nation-states — not just in terms
of the potential economic opportunity, but also from a defensive cybersecurity
perspective — quantum computing could be a technology for which the report’s
framework could apply. Proactively designing and implementing a holistic quantum
computing policy could enable nation-states with sufficient infrastructure —
including those that may not have led the way in the current classical computing
paradigm — to potentially leapfrog their competitors.

We spoke to Celia Merzbacher, Executive Director of the U.S. Quantum Economic

Development Consortium (QED-C), about some of the aspects nation-states should
consider when developing a quantum computing policy.

Expert Interview with Celia Merzbacher, Executive Director

of the U.S. Quantum Economic Development Consortium
(QED-C)
Q: What are the greatest challenges to realizing the potential of quantum
computing?

Celia: There are three main barriers that need to be addressed: The hardware must
be improved and scaled up; new “quantum computer science” is needed, including
algorithms and error correction schemes; and demonstrated use cases must be
established that have substantial market potential. Overcoming these challenges
will require a lot of engineering R&D by the private sector and investment in
fundamental scientific research by the public sector. Identification of use cases
requires engaging end users in depth and can be greatly aided by consortia like
QED-C, which brings together the entire innovation ecosystem.

Q: What lessons can governments that wish to develop their own quantum
computing strategies learn from the development of the quantum computing
industry over the past 10-20 years?
Celia Merzbacher
Celia: A lesson from the early development of the quantum computing industry is
Executive Director, U.S. Quantum Economic
that it is still early! There is not yet a clear winner among the several technologies
Development Consortium (QED-C)
that are being developed — and more than one may be successful for specific
applications. Therefore, governments can assess their nation’s strengths, whether
in theoretical aspects, engineering, manufacturing, or something else, and then
partner and invest strategically.

Q: How should a quantum computing strategy be different than a classical

computing strategy?

Celia: When classical computing was being developed, government (in the U.S. in
particular) was the primary customer and was willing and able to pay a premium to
support that development for government missions, including in the military and
space. Today, computing applications are largely driven by customers in the private
sector, for example finance, chemicals and drug development, logistics,
entertainment, and games. To accelerate the development of quantum computing
that will have applications in both government and commercial sectors, government
should help reduce the risk of early-stage technologies and provide sound export
controls that do not hamper innovation.

Q: To what extent are the leaders of the quantum computing era likely to be
the same nations that have led the classical computing era?

Celia: Nations where there is a robust ecosystem that supports classical computing
have the benefit of infrastructure that could be important for the development and
manufacture of quantum computing systems. However, many advances will emerge
from academic research institutions where researchers have access to the scientific
literature and can collaborate more easily than in the past with colleagues
worldwide. Therefore, breakthroughs — especially in theory and software
engineering, which are less dependent on costly infrastructure and equipment —
could emerge from any corner of the globe.

Corporates
For corporates, the opportunity that quantum computing presents will be far more
specific than for nation-states, so corporates’ actions will be highly specific to their
particular industries and business strategies. From our discussions with numerous
quantum computing companies and the corporate clients they serve, we identified
three key areas that corporates should be aware of when attempting to understand
the quantum computing landscape:

 What forecasts exist for total addressable market (TAM).

 What value creation could look like.

 Why collaboration and the cloud are key to corporate adoption.

Our discussions also showed us that the guidance corporates need depend on their
goals with respect to quantum computing. Hence, we put together two separate
sections with our conclusions for corporates — “How to Prepare: For Quantum
Advantage” and “How to Prepare: For the Quantum Threat.”

Market Forecasts
One broad way of gauging the size of the quantum computing opportunity is to look
at forecasts for its total addressable market (TAM). To do this, we identified 10
reports with perspectives on the market, and using these reports, pulled together a
consensus estimate of the quantum computing TAM through 2027. Based on the
literature, the average market size estimate puts the quantum computing TAM at
just under $650 million in 2022, growing at an average CAGR of around 30% to just
under $3 billion in 2027 (see Figure 46). However, this is only part of the story.

Figure 46. Range of Current TAM Forecasts (2022-27)

($bn)

10.0
9.0
8.0
7.0
6.0
5.0
4.0
3.0
2.0
1.0
0.0
2022 2023 2024 2025 2026 2027

Source: Multiple TAM Forecasts, Citi GPS

There does not seem to be any agreed-upon size for the current TAM for quantum
computing, with 2022 estimates ranging by a factor of 3, spanning anywhere from
around $370 million to $1.1 billion. However, the literature seems to agree the
industry is poised for exponential growth over the current decade, with CAGR
estimates ranging from around 15% on the lower end to more than 50% on the
upper end. Given the large variations in both the current TAM and the growth rate,
this understandably leads to increasingly diverging forecasts (by a factor of over 10)
as we enter the latter part of the decade, as noted in Figure 47.

Figure 47. Current TAM Forecasts (2022-27)

($ mn)

10,000

9,000

8,000

7,000

6,000

5,000

4,000

3,000

2,000

1,000

-
2022 2023 2024 2025 2026 2027
IDC Quince Market Insights
Markets & Markets Brand Essence Research
Global Market Insights BCC Research
Maximize Market Research Verified Market Research
Prescient & Strategic Intelligence Persistence Market Research

Source: IDC, Quince Market Insights, Markets & Markets, BCC Research, Maximize Market Research, Verified Market Research, Prescient & Strategic Intelligence, Global
Market Insights, Fortune Business Insights

Comparing Quantum Computing to Other Areas of Innovation

Of course, in isolation, these TAM forecasts have limited meaning. As part of our
ongoing analysis of the innovation and technology space as a whole, we analyzed
the TAM forecasts for 100+ different areas of innovation, including almost 60
technology-orientated innovations that are often compared to quantum computing.

Our quantitative analysis enabled us to objectively compare different areas of

innovation — something that is otherwise typically very challenging. Our findings
were quite stark, TAM forecasts for various technological disruptions can vary by as
much as 10,000x, with very broad areas of innovation such as Internet Business
Models being several orders of magnitude larger than more narrowly labeled ones
such as Remote Working or Mental Health. Our analysis found that Quantum
Computing was somewhat of an outlier, with one of the lowest TAM estimates out of
all the areas of innovation we looked at.

This makes sense, as the industry is still maturing, and most end-users today use
the technology for research purposes. It also suggests that solely using the
quantum computing TAM to analyze the market opportunity may not necessarily be
the best way forward, which we discuss further in our next section “Value Creation.”

Figure 48. TAM Against 2021-26 Estimated Growth of TAM

E-Aircraft / Air Taxi

60%

5G Network
50%
2021-26 Estimated Growth of TAM (GAGR)

Metaverse
Hyperloop Last-Mile Delivery
AI
40% Edge Computing
Immunotherapy
Experiential Commerce
Voice-Activated Systems E-Vehicles
Virtual Reality Telemedicine
Quantum Computing Neurotechnology
30%
Mobile Payments
Robotic Surgery Cannabis FinTech Healthcare IT
3D/4D Printing Cloud Computing Internet Biz models
Liquid Biopsy
20% Femtech
Sleeptech Smart Grids IoT
Remote Working
Video Games
Digital Identity SaaS
Psychedelic Drugs IP Nanomedicine
Digital Leisure Cyber Security Mobile Devices Demand
10%
Contactless Elder Care
DNA/Genetic
Automation
Mental Health Tech Dark Kitchens
Mobile Network Transition On-Demand Media
Wearables
0%
0 1 10 mRNA 100 Space 1,000 10,000
Deepwater MedTech
Auto Electronics
-10%
Estimated 2021 TAM ($ billions)

Source: Multiple TAM Forecasts, Citi GPS

Our analysis also showed that most areas of technological innovation had broadly
similar estimated CAGRs over the period 2021-26, with a clustering around the 10-
20% range. As can be seen from Figure 48 above, Quantum Computing is
somewhat of an outlier in that it is on the upper end of this spectrum (at an
estimated 31% CAGR), with a similar estimated growth to other innovations such as
Edge Computing and Virtual Reality.

Notably, it has lower estimated CAGRs than other more topical areas of innovation
that have gained mainstream traction in recent years, such as the Metaverse or
Artificial Intelligence (AI). Again, this is not unexpected given the point above about
quantum computers yet to be providing a commercial advantage to businesses. We
anticipate this is likely to change relatively sharply, however, once the technology
reaches the point of quantum advantage — also discussed further in the next
section “Value Creation.”

The above notion of quantum computing being an outlier is further supported when
looking at Figure 49 below — arguably one of the most interesting lenses to
evaluate the quantum computing market.

Figure 49. 2022 VC Investment Against 2026 Estimated TAM

FinTech SaaS
100,000
Clean Tech Mobile Devices Demand
AI
Health & Wellness
Food Innovation
Cyber Security Climate Change
AgTech IoT
10,000 Mobile Payments
3D/4D Printing Automation Cloud Computing
Defense
E-Aircraft / Air Taxi Wearables Video Games
Virtual Reality
2022 VC Investment ($ millions)

EdTech Fossil Fuels Infrastructure

Quantum Computing Dark Kitchens Space
1,000 Indoor Farming Cannabis Hydrogen
Hyperloop Edge Computing Femtech

Mental Health Tech Neurotechnology Mining Capex

100
Psychedelic Drugs
Nanomedicine
Robotic Surgery
Sleeptech
Urbanization
10 Luxury Spend

1
1 10 100 1,000 10,000
2026 Estimated TAM ($ billions)
Source: Multiple TAM Forecasts, PitchBook Data Inc., Citi GPS

When we compared the forecast 2026 TAM to the levels of 2022 VC investment in
private companies, we found that quantum computing was the most densely
invested area of innovation, with an average 2022-VC-Investment to 2026-Forecast-
TAM ratio of over $600 million per $1 billion. For context, the next most densely
invested areas were E-Aircraft at just under $400 million, Agricultural Technology
(AgTech) at just under $300 million, Financial Technology (FinTech) at just over
$250 million, Software as a Service (SaaS) at just under $250 million, and Artificial
Intelligence (AI) at just under $200 million of VC investment in 2022 for every $1
billion of forecast TAM in 2026.

One interpretation may be that VCs are predicting the quantum computing industry
to grow at a much faster rate than current estimates. Another interpretation is that
VCs may anticipate a potentially oligopolistic market structure occurring in quantum
computing, as was the case with classical computing, and are attempting to position
themselves early by investing in some of the key players. We dive deeper into VC
trends in the “Market Participants” section of this report below. Fundamentally, our
analysis allowed us to put the quantum computing TAM and its projected growth
into context. Nonetheless, as we discuss in the next section, the quantum
computing TAM may not necessarily be representative of the whole quantum
computing opportunity.

We spoke to Alex Challans, CEO of The Quantum Insider, about some of the
challenges involved when assessing the quantum computing TAM.

Expert Interview with Alex Challans, CEO of The Quantum

Insider
Q: What are the challenges when trying to determine the realistic total
addressable market (TAM) for quantum computing, compared to other
nascent technologies?

Alex: While we have a reasonable (but not perfect) idea of the potential use cases
of a future quantum computer, there are numerous engineering challenges to
overcome before we start to see devices that will enable this future. Contrary to
some other nascent technologies, there is still significant uncertainty around the
shape of future business models, as well as scarce existing data points
demonstrating current revenue metrics. Most revenue-generating contracts signed
by quantum computing companies today comprise joint research initiatives
underpinned by a heavily consultative element, and they often do not reflect mature
subscription or license revenue models of a mature industry. Indeed, there is still an
important debate around how the revenue or cost savings generated from quantum
use cases should be split between the quantum computing companies and their
Alex Challans end-users. The current TAM estimates are therefore best treated as broad
CEO, The Quantum Insider guiderails.

Q: In your opinion, what is the best way to gauge the TAM for quantum
computing?

Alex: At The Quantum Insider, we saw the high degree of variability in the quantum
computing TAMs already available in the market and wanted to anchor our
estimates to numbers that are publicly available today. We therefore focused on the
quantum computing as a service (QCaaS) market where companies such as AWS
and Microsoft Azure were sharing the pricing of access to quantum computers. This
meant that we could focus on utilization rates and the number of existing live
quantum computers as the key estimated variable. We also liked BCG's way of
looking at "value creation potential," which does a better job of highlighting the
opportunity presented by quantum across different verticals.

While the TAM is an interesting area to investigate for organizations exploring new
markets, we would encourage corporates to avoid getting fixated on precise
percentages and numbers of billions of dollars; instead, they should focus on
inflection or trigger points. By this, we mean proof points and technological
breakthroughs, such as achieving a consistent enterprise use case for an early-
stage (NISQ) quantum computer, which will demonstrate progress toward opening
up significant market potential. Our expectation is that the gradual growth charts
presented in most TAM predictions may end up being more exponential in character
once these proof points have been hit.

Value Creation
An alternative way to gauge the quantum computing opportunity would be to look at
the amount of value creation the technology could provide. To do so appropriately,
however, it is important to be aware of misconceptions in the space.

Recognizing Adjacent Quantum Technologies

Quantum computing often receives much of the attention from the media — so
much so, that, when people hear the word “quantum” they often just assume it is in
reference to “quantum computing.” But, as we have mentioned on a couple of
occasions in this report, quantum computing is just one type of quantum technology.
This presents somewhat of a challenge, as the quantum computing opportunity is
often conflated with the much larger opportunity of quantum technologies, especially
when talking about the broader notion of value creation.

Consequently, despite these adjacent quantum technologies being beyond the

scope of this report, we feel it is important to recognize them. For instance, quantum
communication harnesses the laws of quantum mechanics to protect data and
offers the prospect of creating ultra-secure communication networks, and is
something nation-states are already investigating.128 Quantum sensing, on the other
hand, is a form of advanced sensor technology that can detect changes in motion
and electromagnetic fields, for instance. It offers the prospect of significant
improvements in areas such as measurement and navigation, and quantum sensors
are often described as potentially starting a revolution in a variety of areas from
brain science to air-traffic control.129

Figure 50. Relevant Markets for Quantum Technologies

Sectors Systems Components
Cyber Security Network Equipment Photonics (lasers, photonic crystals, integrated circuits,
imaging, communications)
Brain Diagnostics Spectroscopy MEMS Devices
North Sea Oil Reserves Recovery Accelerometers Gyroscopes and Inertial Measurement Units Cryogenic Equipment
GPS Navigation Devices Sensors
Semiconductors
Source: UK Government Office for Science (November 2016)

There are various synergies in the technical developments between these other
quantum technologies and quantum computing, meaning that, from the perspective
of nation-states, it makes sense to view quantum computing as often just one pillar
of the broader quantum technologies opportunity.130 For instance, the UK
government previously outlined certain sectors, systems, and components in which
quantum technologies represent an opportunity for companies (see Figure 50).131

128 Martin Giles, “Explainer: What Is Quantum Communication?” MIT Technology

Review, February 14, 2019; UK National Quantum Technologies Programme, “UK
Quantum Networks,” accessed July 20, 2023.
129 BAE Systems, “What Is Quantum Sensing?” accessed July 20, 2023; Kai Bongs,

Simon Bennett, Anke Lohmann, “Quantum Sensors Will Start a Revolution — If We

Deploy Them Right,” Nature, May 24, 2023.
130 Ofcom, Quantum Communications: New Potential for the Future of Communications:

Executive Summary, July 28, 2021.

131 UK Government Office for Science, The Quantum Age: Technological Opportunities,

November 2016.

Being aware of these synergies helps put quantum computing into the broader
context of quantum technologies (which itself may have other sizable opportunities
not yet known). At the same time, it is important to note that on a business level, for
corporates, the value creation quantum computing may offer will likely be distinctly
different than those from these adjacent quantum technologies.

The Granular Nature of Quantum Advantage

When looking at the quantum computing opportunity specifically, one of the

challenges with using the forecasts shown in Figure 51 is that all the estimates used
one CAGR value for the duration of the forecast period. However, our discussions
with experts in the quantum computing field (ranging from academics and
governmental bodies to the CEOs of quantum computing companies) indicate it is
unlikely for one CAGR value to apply to quantum computing. This is because value
creation from quantum computing is unlikely to come all at once.

Figure 51. Step Changes in Potential Revenue Arising from New Use-Cases

Source: IonQ

As discussed earlier, instances of quantum advantage will most likely be one of the
key catalysts for commercial adoption and hence, higher potential revenue. Given
the nature of progress in the quantum computing field and the suddenness with
which quantum advantage may manifest itself, numerous inflection points will likely
occur following each specific area of quantum advantage. Figure 51 above provides
an example of such inflection points and the likely step-changes in TAM that may
follow. It is these step-changes, or inflection points, allowing for value creation in
industries that rely on the four areas most impacted by quantum computing:
optimization, machine learning, simulation, and cryptography.

Figure 52 shows how value could be created in areas relating to financial services,
drug discovery, or material design, to name a few. The amount of value creation is
expected to increase as the ability to build larger and more powerful quantum
computers, and run purpose-built applications on these machines, develops.

Figure 52. Value Creation Forecasts Based on a Quantum Computing Roadmap

Source: IBM

Another useful aspect, shown in Figure 52, is how much of quantum computing’s
value creation may be in “other use cases.” As discussed in the UK government’s
The Quantum Age: Technological Opportunities report, not all applications of the
technology may be clear at this point in time or aligned with any particular industry.
IBM estimated there may be over $3 billion in near-term value creation if the
inflection points noted in Figure 52 begin in 2023.132 However, it is almost
impossible to tell exactly where, when, or by how much value creation there will be
in each sector. For instance, BCG estimates the total annual value creation for end-
users in the NISQ era (which they define as pre-2030) as $5 billion to $10 billion.

132 IBM, “The IBM Quantum Development Roadmap,” accessed April 10, 2023.

Figure 53. Potential Use Cases for Four Main Quantum-Advantaged Problem Types

Source: Boston Consulting Group, Citi GPS

Figure 53 provides some relative context to the value creation potential of quantum
computing at the technology’s maturity, which BCG estimates to be around $450
billion to $850 billion in the next 15-30 years.133 Regardless of the precise total
value, the key takeaways are that (1) value creation will likely be divided into the
four broad areas of quantum advantage and (2) each area of quantum advantage
will manifest itself in a highly granular way.

We spoke to William Hurley (better known as “whurley”), CEO of Strangeworks,

about what value creation in quantum computing could look like.

133 Jean-François
Bobier et al., “What Happens When ‘If’ Turns to ‘When’ in Quantum
Computing?” BCG, July 21, 2021.

Expert Interview with William Hurley, CEO of

Strangeworks
Q: How do you think value creation in the quantum computing industry will
manifest itself?

William: The initial value creation will center around efficiencies in computing —
time, cost, and quality. Large-scale problems that can experience marginal
improvements in efficiency can lead to large savings. But the true value creation will
come from problems and applications we haven’t even thought of yet. No one in the
‘50s and ‘60s imagined social media and advertising as a use of computers at the
time. The creation of entire industries is waiting for us when quantum computers
become useful.

Q: Is there a particular area of quantum software that excites you the most?

William: This is admittedly biased, but our work at Strangeworks around the
software infrastructure for quantum computers is incredibly exciting. The quantum
software stack is currently very fragmented. Best-in-class solutions are available
William Hurley
from different hardware and software players. The goal at Strangeworks is to bring
CEO, Strangeworks
all of these together and allow partners and customers to develop services that
draw from all of these capabilities.

Q: Which problems does Strangeworks believe have the most potential for
quantum advantage?

William: I have a slightly different view on this one. Many have suggested things
like quantum chemistry and drug discovery as potential areas from which quantum
advantage will first emerge. I think quantum has much broader applications. As an
example, one of the first areas of quantum advantage we’ve seen is machine
learning experiments. Recently, a team published a white paper on arXiv
demonstrating that substantial quantum advantage can be realized using today's
relatively noisy quantum processors. However, where I see the real opportunity is in
any computational problem where you slightly increase the number of variables and
the evaluation time of the problem increases exponentially. The traveling
salesperson is used a lot as an example, but I believe there are a host of similar
examples across industries.

Q: With more players entering the quantum computing market every year,
what will give some players a competitive advantage?

William: Players that can effectively educate the customer about the promise of
quantum computing as it relates to their industry and existing workflow will gain
traction more quickly than those waiting for industries to come around. Many
potential customers begin by asking us what the potential use cases are in their
industry vertical — even those with quantum computing teams! The challenge is to
get past the decision maker and connect to the subject matter expert or end user,
discover their bottlenecks, and give them the resources to become advocates in
their organization.

Collaboration and the Cloud

We feel that collaboration and the cloud will be key to the corporate adoption of
quantum computing.

Understanding the Need for Collaboration

As we touched on in the “Nation-States” section at the beginning of this chapter,
collaboration is likely key in developing a quantum computing ecosystem.
Analogous to how building quantum computers will require collaboration between
experts in fields such as physics and computing, commercializing quantum
computing for corporate adoption will require collaboration between quantum
computing companies themselves to build an efficient infrastructure — and most
importantly, one that enables corporates to extract value.

Quantum computing companies will likely need to work together to standardize the
technology. One of the easiest ways to understand this is to consider the role
collaboration has historically played. For example, it was only with the advent of the
universal serial bus (USB) protocol — a standardized connector agreed upon by
industry players — that computer peripherals could be moved efficiently from
machine to machine. Removing this friction allowed for a focus on the consumer
experience.

Hence, on top of the needed collaboration between the academic and private
sectors (as much of the theory behind quantum computers comes from academia),
collaboration between quantum computing hardware and software companies will
accelerate the path to quantum advantage. This is in part because, for many of the
complex quantum algorithms and programming languages to run efficiently, they will
need sufficient error-correction, which itself is an engineering problem that needs to
be addressed in conjunction with the manufacturers of quantum computing
hardware.

Figure 54. Breakdown of Layers Between Quantum Computing Hardware and Algorithms

Quantum Algorithms and Programming Languages

Error Correction

Quantum
Photons Superconductors Ions Semiconductors
Dots

Quantum Computer Hardware

Source: UK Government Office for Science (November 2016)

The Role of the Cloud

A number of competitors have begun offering access to their quantum computers
via the cloud, dubbed Quantum Computing as a Service (QCaaS). There was
consensus among experts we spoke to that quantum computing will mostly be a
cloud-based service in the immediate to medium-term future. And in that sense,
quantum computing’s starting point will likely continue from classical computing’s
apparent endpoint.

The cloud is expected to accelerate the adoption of quantum computing, reducing

the costs and challenges faced by corporates to maintain their own infrastructure
and have a quantum computer on-premise (for example, the need for refrigeration
and a team of engineers to maintain the quantum computer). A report by The
Quantum Insider estimated the QCaaS market to be less than $50 million in 2020
and to consist mainly of research projects, consulting projects, and experimental
access to platforms such as IBM Quantum and Amazon Braket.134 They expect the
QCaaS market to achieve an 80% CAGR over the forecast period of 2021-30. They
estimate the TAM reaches $4 billion by 2025 and $26 billion by 2030, driven by
what they refer to as “the realization of useful applications starting to emerge
midway through this decade.” As Figure 55 below shows, however, the revenue
generated by QCaaS providers is expected to only form one part of the overall
quantum computing TAM.

Figure 55. Breakdown of Different Layers Between End Customers and QPU Providers
Excludes Layers of Enabling Hardware and Software

Software/ QCaaS Provider QPU Provider

End Customer
Application Layer (e.g., AWS) (e.g., IonQ)

Revenue generated by
providers of access to
QCs only one part of
the total quantum
computing addressable
market

Source: The Quantum Insider135

QCaaS providers are uniquely positioned between the software layer and the
Quantum Processing Unit (QPU) to identify trends in corporate quantum computing
usage. Using quantum computing cloud services also provides the end-user the
flexibility to access different types of quantum computing technologies, such as
quantum annealing or gate-based quantum computers, depending on their needs at
the time. This is likely to play a particularly important role in the coming years, as
corporates investigate different use cases and potentially experiment with different
qubit technologies, ahead of the industry achieving quantum advantage (which we
describe earlier as the point at which QCs can offer practical advantages in solving
a valuable problem, whether that is by enabling faster, cheaper, or more efficient
solutions than classical computers).

We spoke to Professor Simone Severini, Director of Quantum Computing at

Amazon Web Services (AWS), about the role collaboration and the cloud will play in
the progress and adoption of quantum computing.

134 The Quantum Insider, “Report: Quantum Computing as a Service Market to Hit $26
Billion by End of Decade,” August 12, 2021.
135 The Quantum Insider, “Quantum Computing as a Service Market Sizing – How We

Did It,” August 19, 2021.

Expert Interview with Professor Simone Severini, Director

of Quantum Computing at Amazon Web Services (AWS)
Q: What role will collaboration with hardware providers play in the adoption of
quantum computing?

Prof. Severini: AWS has the largest and most dynamic community, with more than
100,000 Partners from over 150 countries. From a quantum standpoint, hardware
providers are critical for expanding access to quantum computing through Amazon
Braket, our managed quantum computing service. With Braket, we are giving our
customers access to a variety of approaches when it comes to trying quantum
hardware — all delivered through our hardware partners.

We believe there is no “right” path to exploring quantum. The truth is, it is too early
to know which approach will become the standard, which is why we are focused on
providing access to a range of different approaches, such as gate-based ion-trap
processors, superconducting processors, and photonic quantum computers.

Q: How does the cloud enable collaboration in the quantum computing space
Professor Simone Severini
more generally?
Director of Quantum Computing, Amazon
Web Services (AWS) Prof. Severini: First and foremost, the cloud enables access. Customers are
choosing AWS over other providers because it has a lot more functionality, the
largest and most vibrant community of customers and partners, the most proven
operational and security expertise, and the business is innovating at a faster clip —
and this extends to AWS’ quantum computing portfolio and services.

The magic of Amazon Braket is that it’s running on AWS, where our customers data
already lives, alongside storage, analytics tools, and all of the other services our
customers are already using to run their business. Amazon Braket is fully integrated
with AWS, providing a seamless experience for customers to leverage quantum and
classical resources together for the first time. This will be critically important in the
future where classical and quantum will work together to solve problems.

Q: What are the advantages of quantum computing on the cloud vs on-

premises?

Prof. Severini: There are five reasons companies are moving so quickly to the
AWS cloud, and by extension, Amazon Braket: (1) agility, (2) cost savings, (3)
elasticity, (4) the ability to innovate faster, and (5) the ability to deploy globally in
minutes. Amazon Braket is just like any other of AWS’s services, and our customer
experience reflects that. By running on AWS, our customers will always have
access to the latest hardware vs. an on-premises system that could potentially be
out of date before it’s even implemented. In addition, they don’t need to worry about
reconfiguring their data centers, or worse — having to manage such delicate
machines and the overhead that comes with it.

Q: What are the most important technological developments needed for the
continued adoption of quantum computing on the cloud?

Prof. Severini: Quantum computing is still in the very early stages of research. With
that in mind, the primary technological development that needs to be improved is
error rate, or how accurately we can perform quantum gates. Quantum devices
available today are noisy and are as a result limited in the size of circuits that they
can handle (a few thousand of gates is the best we can hope for with NISQ
devices).

This in turn severely limits their computational power. There are two ways that we
are approaching making better qubits at the AWS Center for Quantum Computing:
(1) by improving error rates at the physical level, for example by investing in
material improvements that reduce noise; and (2) through innovative qubit
architectures, including using Quantum Error Correction to reduce quantum gate
errors by redundantly encoding information into a protected qubit, called a logical
qubit.

How to Prepare: For Quantum Advantage

With the cloud providing a means of access to quantum computing, and with many
quantum computing companies mapping out when and where they feel quantum
advantage will occur this decade, the question turns to what corporates can do now
to capitalize on quantum advantage when it occurs.

Figure 56. The Path to Quantum Advantage

Quantum Quantum Quantum

Awareness Preparedness Advantage

Source: Citi GPS

Phase 1: Quantum Awareness

Our conversations with both corporates and industry experts revealed that while
there is awareness of quantum computing among corporates, it is often considered
a distant future technology or associated with previously-promised unrealistic
timelines in delivering an advantage over current classical computing methods.

To that end, there are a lot of misconceptions about what quantum computers can
and cannot do, and often little awareness of the specific areas that quantum
computing will have an impact. It is also important to be aware of the tangible
impact quantum computing may have on the particular industry or industries that a
corporation operates in.

Phase 2: Quantum Preparedness

Given the diversity of corporations, not only in the industries in which they operate,
but in their size, budget, and ambitions, no two corporates are likely to follow the
exact same path toward quantum advantage. Figure 57 illustrates a general five-
step process that companies can look to in preparing for the age of quantum
advantage.

Figure 57. Five-Step Plan to Prepare for Quantum Advantage

Contact Create Your Identify an Plan to

Look Your Quantum Appropriate Integrate
Internally Existing Impact Quantum Quantum
Partners Assessment Partner Computing

Source: Citi GPS

Step 1: Look internally Given that quantum computing is a complicated topic, spans numerous academic
disciplines, and faces a shortage of specialists, companies will need to identify a
primary point person to deal with outside partners and address workforce education
requirements. As identified earlier in this report, talent shortages and upskilling the
existing workforce will also be challenges. One solution may be to train employees
to read technical papers, to help them better understand and decipher technical
claims when they are made.

Step 2: Contact your existing partners It is important to look among existing partners and identify individuals with specific
quantum expertise. These may be IT partners, security vendors, cloud providers,
research houses, or consultants. Existing partners will each have their own domain
expertise in how a business operates, as well as their own unique perspective on
where value in quantum computing may first occur. They may also help to identify
the operational weak links which could be susceptible to quantum attacks.

Step 3: Create your quantum impact Armed with both the expertise of an internal workforce and existing strategic
assessment partners, the impact quantum computing may have on your business can be
gauged. This includes identifying specific problems and opportunities and
determining their materiality. It should also involve creating a business timeline and
ideally should include discussions with the C-suite and at the board level.

When putting together a quantum impact assessment, it is important to be aware of

what existing corporate partners do not know. Many will likely consider quantum
computing as still in its infancy and will not necessarily be prepared for it or be able
to provide guidance on the future services required. As such, depending on the
knowledge gained from the two prior steps, it may be appropriate to swap Step 3 in
Figure 57 with Step 4 to gain an agnostic view on the impact quantum computing
may have on a corporate.

Step 4: Identify an appropriate quantum Identifying the kind of partner needed is complicated: Some quantum computing
partner companies focus primarily on providing access to hardware, some operate only on
the software side, and others aim to offer a full-stack solution across all elements.
Each provider will notably have their own limitations in terms of services they can
provide. For instance, some full-stack quantum computing providers may not be
agnostic to the hardware of competitors. Alternatively, software providers may be
biased toward using the hardware modality they specialize in. And even within these
different areas of focus, there is considerable variation among quantum computing
companies in terms of their expertise and services. There are numerous criteria to
examine in this regard, but some questions to ask may include:

 How deep are they in the science?

 How much can they integrate with my company’s existing enterprise

architecture?

 Can they work across multiple hardware types?

 What types of problems can they solve and on what time horizons?

For instance, in terms of hardware, gaining access to quantum annealers in the

future may be particularly beneficial if capitalizing on quantum advantage in the
area of optimization is a priority. On the other hand, gate-based quantum computers
may be better if future plans involve more general-purpose quantum computing
applications such as machine learning or molecular simulations. Be aware,
however, that the two types of technology are at different stages of development. In
terms of software, the availability of different quantum computing partners reflects
the diversity of the industries that quantum computing will impact. Some software
providers aim to provide a general-access software solution, whereas other
providers will focus on providing the optimal software package for a specific
industry, such as finance.

Something else to consider when identifying a quantum computing partner is how

fast quantum computing is expected to be integrated within business workflows and
infrastructure. From speaking to industry experts, this can vary significantly, and
take anywhere from 12 to 36 months. One of the key questions to ask at this stage
centers around the motivation for adopting quantum computing technology. Is it to
secure an early competitive advantage ahead of the expected exponential increase
in quantum computing capabilities? Or is it to ensure not being left behind when
competitors eventually adopt this technology?

Step 5: Plan to integrate quantum computing After identifying the appropriate quantum computing partner or partners (many
corporates are opting to engage with more than one), the focus should shift to
incorporating a means of access to quantum computing into the existing tech stack.
This is unlikely to be “plug and play,” given that the foreseeable future is a hybrid
one in which corporates will need both quantum and classical computing depending
on the task at hand.

Part of the integration process is to start building solutions to actual business

problems that will force an integration into existing classical computing systems.
Engaging in a “proof of concept” in this manner will allow for a better determination
of a long-term quantum computing strategy. Notably, however, the proof-of-concept
stage is often constrained to a research environment, so a true pilot would need to
use real data and operate in the constraints of the particular business unit. Only
then could one truly test the feasibility of outperforming existing classical-only
solutions.

Engaging in such a process should mean that corporates are not caught off-guard
by a lack of technical integration or skills within teams when the time comes to
benefit from quantum advantage.

Phase 3: Quantum Advantage

Notably, the path to quantum advantage will not necessarily be easy or cheap —
especially for those playing to win. There are two key levels at which to explore
adopting quantum computing. The first is simply to investigate and become familiar
with a platform; for example, gaining access to a low-qubit quantum computer to
understand how quantum algorithms work or may fit into your workflow. The second
is to invest significantly in the technology and obtain access to a higher-qubit
quantum computer to run the cutting-edge algorithms needed to position yourself for
quantum advantage.

In 2022, Zapata Computing commissioned their second annual survey of 300 senior
executives at large global enterprises with estimated revenues of over $250 million and
computing budgets of over $1 million.136 They found that 71% of quantum-adopting
enterprises had a quantum computing budget of $1 million or more — considerably
higher than in 2021 (28%). Also, 33% of respondents were what Zapata classified as
“early or most advanced” tech adopters. Additionally, the percentage of respondents
indicating they had no plans to adopt quantum computing fell from 31% in 2021 to 26%
in 2022.

We spoke to Dr. Christopher Savoie, CEO of Zapata Computing, about their findings
and what they are seeing from corporates today.

136
Zapata Computing, The Second Annual Report on Enterprise Quantum Computing
Adoption, January 11, 2023.

Expert Interview with Dr. Christopher Savoie, CEO of

Zapata Computing
Q: What are the earliest use cases that could provide an advantage for
quantum computing over classical computing?

Dr. Savoie: Zapata believes the fastest path to quantum advantage could be with
quantum-enhanced generative AI. Here, quantum computers are used as sampling
devices to generate candidate solutions that classical computers cannot generate.
We’ve used this approach to generate new solutions for optimization problems, for
example optimizing assembly line scheduling as we recently demonstrated in work
with BMW. Drug discovery is another promising near-term use case. We just
published a study with Insilico Medicine and Foxconn that showed quantum-
enhanced generative models can outperform classical generative models at
generating small molecules. But we don’t have to wait for quantum computers to
mature. The same quantum methods can run on current classical devices to
outperform best-in-class algorithms today — albeit, with less power than large-
scale, error-mitigated quantum computers will provide.
Dr. Christopher Savoie
CEO, Zapata Computing
Q: What is the role of classical computing with quantum applications?

Dr. Savoie: Quantum solutions will likely involve mostly classical computing, with a
few powerful quantum computing steps. This includes processing data before it
goes in the quantum algorithm as will any post-processing needed to turn the
quantum outputs into actionable insights. Applications must orchestrate across the
quantum and classical systems in the context of complex, hybrid-cloud, customer-
specific architectures. At Zapata, we’re solving for all the complexities of classical
analytics and quantum in one integrated solution.

Q: What are the greatest barriers organizations face when adopting quantum
computing?

Dr. Savoie: The largest barriers are the complexity of integrating quantum into the
existing IT stack and the shortage of quantum talent, which is much scarcer than
data science talent. But more importantly, customers are not clear on which use
cases are most viable near-term. Some ecosystem players promote use cases that
are helpful to customers only as research proofs of concepts (POCs): For example,
variational quantum eigensolver (VQE) for molecular simulation is 7-10 years from
outperforming current methods. The truth is that quantum-enhanced generative AI
can add value today and should be the first place enterprises look for use cases.

How to Prepare: For the Quantum Threat

As discussed earlier in the “Cybersecurity” and “Cryptocurrency” sections, quantum
computing poses a risk to existing industries. One report estimated the value at
stake in finance, automotive, pharmaceutical, and chemical use cases to be
potentially approaching $700 billion by 2035.137

As investigated in detail in earlier chapters, quantum computers (QCs) will excel at

a specific set of problems, some of which push the boundaries of math, science,
and technology. They will also be exponentially better at solving problems involving
linear algebra, prime factorization, and discrete logarithms — keys to modern
cryptography. Since much of the world’s current cryptographic infrastructure relies
on classical computers’ inability to solve these problems efficiently, the advent of
quantum computing presents a significant threat. This section intends to expand on
our work in Chapter 2 to outline the problem and address what large organizations
can do right now and in the future.

Understand Symmetric Encryption and the Threat to It

Think about encryption like a lock on a door. Many locks have been designed, but
over time, a particular shape and style of lock is used fairly universally on all doors.
This “standard” lock format represents an encryption algorithm, which is the
mathematical function that “locks” or “unlocks” the metaphorical door. In this
metaphor, the security of the lock is primarily based on each lock requiring a
different key, not a different lock design. The key for the lock is like an encryption
key. Since a locksmith or burglar cannot see inside of a lock to know which key will
fit, they use special tools like lockpicks to “feel” inside the lock until they work out
the shape of the correct key. Once they are successful, the lock opens.

Modern cryptography and attacks against it are similar. Although the locking
mechanism is known via published algorithms, the security of the system is based
on maintaining the secrecy of the encryption key. To bypass the encryption, an
attacker must try many keys to find the one that fits or attempt to figure out what the
lock expects and create a key that will open the door.

In symmetric encryption algorithms like AES (Advanced Encryption Standard) and

its predecessor DES (Data Encryption Standard), the same encryption key is used
to encrypt and decrypt through the same algorithm. In our door lock analogy, the
locksmith is going to need to work out what key will fit with almost no information
about the shape of it. They will need to test lots of possible keys until one fits. This
type of attack, known as brute-force, creates a random encryption key value which
is then used to attempt to decrypt the data.

If our lock only had a couple of pins on it, then working out the right key would be
easy. The same is true in symmetric encryption. If there were only a small number
of potential encryption keys, even into billions or trillions, then a traditional computer
would eventually be able to guess the right key. This concept is known as “key
size.” The larger the key size, the more difficult it is for an attacker to guess the right
key. AES, for example, has key sizes of 128, 192, and 256 bits available. If we use
a 128-bit key, there are around 340 undecillion unique keys that would need to be
attempted to brute-force decrypt the data — that is 340 with 36 zeros after it! Even
leveraging the fastest modern classical computers in the world would take millions
of years to brute-force that key.

137
Matteo Biondi et al., “Quantum Computing Use Cases Are Getting Real—What You
Need to Know,” McKinsey, December 14, 2021.

In 1996, Lov Grover devised an algorithm to run on quantum computers making it

significantly easier to perform an unstructured search. That is, his algorithm could
determine much more efficiently what unknown value should be fed into a
mathematical function to produce a particular output. This obviously causes some
concern around the continued viability of symmetric key encryption, which relies on
the difficulty in determining which key in the massive potential pool is correct.

However, Grover’s algorithm only provides about four times the speed of a
traditional searching algorithm. So, systems like AES are not completely broken.
The number of potential keys available simply need to increase to maintain the
security of symmetric algorithms. Many organizations are suggesting the adoption
of 256-bit keys as a standard to protect against attacks from Grover’s algorithm.
This is based on the understanding that today, a 128-bit key is considered safe for
protecting sensitive information up to the Top-Secret level, and doubling the key
length is considered sufficient to provide the same level of protection in a post-
quantum environment.

Understand Asymmetric Encryption and the Threat to It

Imagine a treasure chest that is locked with one key but can only be unlocked with a
different key. A pirate could store away some treasure for their buddy and be sure
that only the person who has the unlock key can retrieve it. Cryptographers would
call this system an asymmetric (or two-key) algorithm — also known as public-key
encryption. In these algorithms, there is a public and private key pair that are
mathematically linked. So, our locksmith here could analyze the public key to
determine the private key, in theory. Public-key encryption (such as RSA, discussed
earlier in the report) makes this difficult by using mathematical features like prime
factorization to make that analysis incredibly difficult for traditional computers to
solve.

Peter Shor, a MIT mathematics professor, published a paper in 1994 that outlined a
quantum algorithm for solving a set of problems that traditional computers found
exceedingly difficult. His algorithm “solves” for prime factorization — in other words,
it can find the prime numbers that, when multiplied together, give a particular value.
Since asymmetric algorithms have a public part of their key, Shor’s algorithm can
effectively determine the private key part very quickly. The mathematical
relationship between the two key pieces means that current asymmetric algorithms
can be considered broken once Shor’s algorithm can be run against large enough
numbers. Simply increasing the key size will not fix this issue; new quantum-safe
algorithms will be required.

Recognize the Threat Timeline

Cryptography underpins the security of every computer system we interact with.

Symmetric algorithms protect our data in storage, in backup tapes, on disks, and
over the network. Asymmetric algorithms are used for establishing secure network
connections (like SSL, TLS, and VPN), maintaining strong identities, and
exchanging symmetric encryption keys. Generally speaking, asymmetric algorithms
protect data in transit while symmetric algorithms protect data in storage. Web
security, digital identities, encrypted phones and laptops, web servers, encrypted
databases, and more would all be fundamentally broken by quantum
implementations of Grover’s and Shor’s algorithms.

While Grover’s and Shor’s algorithms are known, they cannot be run on today’s
quantum computers in a way that would significantly impact our current encryption
systems. However, there are a few problems. First, the point at which the required
size of quantum computers will be ready is unknown. Current estimates from
“friendly” companies and researchers suggest anywhere from five to 10 years, but
there could be significant advances by our adversaries that we do not know about.
Also, our encrypted communications and data are likely already being archived by
our adversaries for a future date when they can be decrypted — known as a
“Harvest Now, Decrypt Later” (HNDL) attack. If any of our data could be exploited in
five to 10 years, then we need to take steps to protect it today.

Develop an Enterprise Approach to Cryptography Modernization and

Agility

We see two main approaches to cryptography, which can be done in parallel:

cryptography modernization and cryptography agility.

Single-key (symmetric) encryption systems that are in use today will need larger
keys to ensure they stay effective once quantum computers are available. Current
standards and implementation need to be enhanced to establish a 256-bit key
length as the default for AES and use similar-strength keys for other symmetric
encryption systems. This will ensure today’s encrypted data cannot be broken in the
future.

When current public-key encryption algorithms can be completely broken by

quantum computers, those algorithms will need to be replaced with quantum-safe
solutions, known as Post-Quantum Cryptography (PQC). NIST and other
organizations are already determining which new algorithms should be adopted.
Once available, those new algorithms will need to replace today’s public-key
encryption in every platform that uses it. This will be a huge undertaking, similar in
scope to the Y2K problem, as every technology developer and company will need to
participate.

For some corporates, the best approach may be to start working on a plan that will
allow quick and easy identification of which systems are impacted, develop ways to
swap out algorithms efficiently, and collaborate with vendors and partners to ensure
their systems are also updated. This will become even more difficult when
corporates move to the stage of using End of Life and End of Vendor Support
systems (e.g., older versions of computers that do not receive the latest software
patches), as these will not likely be patched to support new algorithms. The same is
true for traditional cryptographic protocols that have been superseded — they will
almost certainly not support the new encryption paradigms required.

Regardless of the approach taken, for most large corporates, cryptography

modernization will require a multi-year strategy that will impact policies and
standards and potentially lead to technology program changes and partnerships.

Quantum-resistant algorithms are being developed on today’s classical computers,

meaning one does not need a quantum computer to design quantum-resistant
algorithms. However, there are also quantum cryptography algorithms that could be
adopted in the future that, by design, would be quantum-proof. A determination on
how widely these technologies are adopted is necessary before deciding on
implementing quantum cryptography.

NIST began formally investigating PQC in 2016. In addition to knowledge-sharing

and presentations around the topic, their primary focus has been in identifying
algorithms to adopt and replace those identified as vulnerable and is expected to
release draft standards. Given the preexisting shortage of cybersecurity experts,
companies need to be mindful of the likely talent shortage in quantum computing as
well, as discussed in the “Workforce Education” section of this report.

We spoke to Professor Deeph Chana, Chair of the NATO Advisory Group on

Emerging and Disruptive Technologies, about what organizations should consider
when it comes to the threat of quantum computing.

Expert Interview with Professor Deeph Chana, Chair of the

NATO Advisory Group on Emerging and Disruptive
Technologies
Q: What should organizations take into consideration when evaluating the
quantum threat to cryptography?

Prof. Chana: We can say with some confidence that the current situation is
characterized by a significant lack of technical literacy and competence within the
majority of organizations where security issues related to quantum technology
should be of concern. Such skills are simply hard to attain and are in scarce supply.

Furthermore, well before we get to the highly advanced topic of quantum

computing, we already have evidence, through countless studies and real-world
events, that significant issues persist with respect to the base competencies needed
to deal with more conventional forms of cybersecurity risk, not to mention the
threats that might be posed by a quantum computer. That this is a global issue has
been revealed by recent incidents such as the WannaCry ransomware attack, the
SolarWinds supply-chain software hack, the myriad of cybersecurity issues related
Professor Deeph Chana
to the COVID-19 pandemic, and the modus operandi in cyber-warfare used in the
Chair, NATO Advisory Group on Emerging
Russia-Ukraine war.
and Disruptive Technologies
In each case, governments and industry leaders appeared to be caught by surprise,
in many cases claiming, incorrectly, that such risks were previously unknown when
in reality they had been openly highlighted by experts, in some cases for over a
decade. In the short term, therefore, it seems that most companies will have to rely
on external expertise to understand the importance of quantum computing risks.
Ideally, however, their needs will be best served by a blend of internal competence
collaborating with external expertise, and this is what they should plan for.

Q: Why should organizations be thinking about future quantum attacks right

now?

Prof. Chana: It is not necessary for quantum computing to be a commercial reality

in order for considerations of its security risk potential to be undertaken. More
importantly, with such an analysis in hand, it is feasible to start taking action through
pre-emptive implementations of technical and non-technical mitigations. Security
preparedness through designed resilience rather than a policy of reaction, therefore,
is the posture I’m advocating for.

However, related to my points on expertise above, this requires competent

technology risk analysis to be well established and available to an organization,
which, currently, is generally not the case. The major issue with not taking a stance
of preparedness may be summarized by the idea of accumulation of security debt;
where the impact of a quantum computing-driven threat would be greatly multiplied
in future systems that evolve and grow in scale, complexity, and reach without
designed-in mitigations.

We can, for example, consider scenarios where large amounts of conventionally

encrypted data might suddenly become accessible due to the arrival of a quantum
computing-enabled decryption system — this should be motivation for taking action
on advanced encryption and data security methods now. The idea of stealing data
today for the purposes of decrypting it tomorrow is a threat scenario that needs
consideration across a range of industries.

Market Participants
Overview
This past decade has seen increasing amounts of private investment into quantum
computing. As it is no longer just an esoteric technology of governments and
university departments, private markets are taking note of the increasing number of
spin-offs and start-ups making inroads in the area.

While there are, of course, examples of consolidation occurring in the space, there
is not yet a consistent level of M&A activity, which is why we focus on the capital
invested in the private primary markets.

The two broad areas market participants need to be aware of are (1) venture capital
(VC) trends and (2) the company and funding environment.

Venture Capital Trends

VC Investment

The dollar amount of capital invested in quantum computing has increased sharply
in recent years. 2020 seems to have been a breakout year for VCs in the sector,
with over $700 million invested, almost as much as the five prior years combined
(Figure 58)

While some may attribute this to the so-called “spec-tech boom” following the global
interest rate cuts in light of the COVID-19 pandemic, 2021 saw this trend continue
with over $1.3 billion invested (Figure 58), accounting for almost half (around 44%)
of all the capital ever invested by VCs in quantum computing at the time.

Figure 58. VC Investment in Quantum Computing (2010-1Q 2023)

($bn)
2.0 Angel and Seed Early-stage VC Later-stage VC

1.5

1.0

0.5

0.0
2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 1Q
'23

Source: PitchBook Data Inc., Citi GPS

Another potential catalyst for investment in recent years could have been the claims
of quantum supremacy in 2019, which brought quantum computing to the attention
of the public. In the three years leading up to this announcement, VC investment in
quantum computing saw only a 29% CAGR, rising from $105 million in 2016 to
$227 million in 2019, whereas the three years since has seen a CAGR of more than
150%.

Despite the market turmoil of the first half of 2022, quantum computing received
strong investment of nearly $1.8 billion (Figure 58), equating to around a third of all
VC investment to date. In fact, 80% of the VC investment in quantum computing
has occurred since the start of 2020, and 95% has been since the middle of the
past decade. Looking at VC investment by stage over the period starting from 2016,
we find that 9% was categorized as angel and seed investment, whereas 41% was
early-stage investment and 50% was later-stage investment. Looking at the average
deal size by investment stage also shows that deals have grown across all
investment stages in recent years, but particularly in the more mature end of the
investment spectrum, such as later-stage investment.

Assessing a nascent technology such as quantum computing by looking solely at

levels of VC investment growth on its own may not tell the full story, especially given
the small starting levels of VC investment in the quantum computing industry. One
technology, however, that quantum computing is often compared to in terms of its
potential impact is artificial intelligence (AI).

Figure 59. VC Investment in Quantum Computing vs. Artificial Intelligence (2010-22)

($bn)

120

100

0
2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022

Quantum Computers Artificial Intelligence

Source: PitchBook Data Inc., Citi GPS

When compared to AI, based on PitchBook data, we find that the level of investment
in quantum computing is very much still a drop in the ocean, with AI having
averaged almost 90 times more investment than quantum computing in the past two
years, with $66 billion and $113 billion invested in 2020 and 2021, respectively. The
recent decline in AI investment $71 billion in 2022, per PitchBook data, alongside
increasing capital flowing into quantum computing as the industry matures, puts the
multiple nearer to 40.

While AI is by no means a perfect comparison to quantum computing, it does show

that, despite quantum computing being one of the most densely invested sectors
(see the “Market Forecasts” section), there is considerable scope for further
investment flows.

VC Deal Activity

In contrast to VC investment, VC deal activity in quantum computing has been on a

steady rise for most of the past decade. Looking at deal count by year, for instance,
deal activity in the space started to take off in the middle of the past decade —
climbing from an average of five deals annually for the period 2010-15 to around 65
in 2020-21. We also see that many of the deals are still in the Angel and Seed
stages (reflecting the nascency of the quantum computing sector as a whole),
although recent years show a trend towards more typical Early-Stage and Later-
Stage investments. 2022 was another record-breaking year, with 81 deals,
suggesting that more players are entering the maturing market.

Figure 60. VC Deal Activity in Quantum Computing (# of Deals, 2010-1Q 2023)

90
Angel and Seed Early-stage VC Later-stage VC
80

2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 1Q
'23
Source: PitchBook Data Inc., Citi GPS

Regional Divides

In terms of regions, the North American and European markets accounted for 73%
of all deal activity over the ten-year period from 2012 to 2021 (the first two years of
the 2010s had so few deals that the data on regional divides for that period is
unlikely to be meaningful). Since that time, however, when looking at the capital
invested, despite the North American market having had 1.3 times as many deals
as the European market, it has had 3.7 times the amount of capital invested.
Combined, the North American and European markets have been responsible for
around 80% of the total capital invested by VCs in quantum computing worldwide
over the past decade.

Figure 61. VC Investment in Quantum Computing by Region Figure 62. VC Deal Activity in Quantum Computing by Region
100% 100%

80% 80%

60%
60%
40%
40%
20%
20%
0%
0%

North America Europe Asia Middle East Oceania

Note: 2023 figures through end of 1Q Note: 2023 figures through end of 1Q
Source: PitchBook Data Inc., Citi GPS Source: PitchBook Data Inc., Citi GPS

As we can see from Figure 61 and Figure 62, Asia has made notable progress in
recent years. According to PitchBook data, 2015 was the first year in which the
region had more than one deal. Since then, deal activity has increased to an all-time
high of 19 in 2021, accounting for 26% of all deal activity. Deal activity has been a
leading indicator for the amount of capital invested, as the VC capital invested in
Asia grew from just over $13 million in 2019 to just over $400 million in 2021.

We spoke to Stuart Woods, Chief Operating Officer of Quantum Exponential, about

some of the VC trends he is seeing in quantum computing.

Expert Interview with Stuart Woods, Chief Operating

Officer of Quantum Exponential
Q: What do you think will be the main driver of VC investment in quantum
computing over the next decade?

Stuart: The incentives for VC investing in quantum computing have evolved quite
significantly over the last few years. Initially, investment was in pure quantum
computing as it related to encryption, and then it was quantum computing for
“supremacy” over classical alternatives.

It may be in the middle- to long-term future that quantum technologies are

integrated into the fabric of much of our infrastructure, but not necessarily fronted by
the quantum buzzword. They will just be services, accepted as standard, that
happen to use and rely on quantum technologies. In the near term, I think installing
quantum computers in data centers for their revenue-generating new services will
become ever more familiar.

Q: What do you think will be the impact of the recent downturn in markets in
Stuart Woods
terms of raising funds for quantum computing companies?
Chief Operating Officer, Quantum
Exponential Stuart: The first key point is that quantum technology is not just quantum
computing. The current global geopolitical environment highlights even further that
quantum is a much broader market than the computing element for which it has
attracted most attention. Quantum sensing, for example, will have varied
applications across timing, GPS and navigation systems, brain imaging, and
measuring ground movement and subsidence. The effect of the downturn is to
focus the market more broadly on other applications needed by governments, for
example, smart infrastructure, clocks for timing financial transactions and sensing
applications for government climate change policies.

Q: What makes investing in quantum computing different from investing in

other areas of Deep Tech?

Stuart: Quantum is more technically challenging, although it is becoming

increasingly mainstream and a must-have asset class. I am wary of parts of the
market that are labelled “quantum-inspired.” This is a misnomer which is highly
ambiguous and has the potential to distort the actual standing and successes of
companies in the quantum ecosystem.

Naturally, technical due diligence and robust intellectual property are key. I also
think understanding the channel to market with quantum is essential; almost all
quantum activities can be spun up into services, but sometimes founders
(misguidedly) want to — or feel they should — sell products. Similarly, not every
quantum company needs to build a quantum computer. We have to remember that
and encourage each facet of the market to nurture and fine-tune its areas of
expertise — not everyone can do everything all at once.

The Company and Funding Environment

It is not just in VC investment and deal activity that a geographical divide exists, but
also in the number of quantum computing companies headquartered by global
region. When looking at company count by geography, we found that, as of the end
of 2022: 45% of companies were headquartered in North America and 34% were in
Europe, followed by 14% in Asia and 3% in the Middle East.

Figure 63. Quantum Computing Company Count by Region (as of end-2022)

Source: PitchBook Data Inc., Citi GPS

Despite the increase in both VC investment and deal activity over the past decade,
according to PitchBook data, the number of quantum companies founded over time
has varied considerably. From an average of three companies founded per year at
the start of the decade to a peak of just over 50 in 2018, there has been a steady
decline over the past few years. For context, 2022 saw just 16 companies founded.

Figure 64. Number of Quantum Computing Companies Founded (2013-1Q 2023)

0
2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 1Q '23

Source: PitchBook Data Inc., Citi GPS

Figure 65 gives some examples of recent VC deals within the quantum computing
sector listed on PitchBook. It is clear that capital is being raised by quantum
computing companies across the globe.

Figure 65. Largest Quantum Computing VC Deals in 2Q 2023

Deal Size (US$) Country Deal Type Deal Date
51.65 United Kingdom Later Stage VC April 18, 2023
30.00 Germany Later Stage VC June 22, 2023
14.00 Finland Early Stage VC June 2, 2023
5.83 Denmark Early Stage VC June 1, 2023
5.81 France Early Stage VC May 23, 2023
5.42 Netherlands Later Stage VC April 13, 2023
4.52 Denmark Later Stage VC May 11, 2023
4.00 United States Seed Round June 6, 2023
3.86 South Korea Early Stage VC June 28, 2023
0.85 Sweden Early Stage VC June 20, 2023
Source: PitchBook Data Inc., Citi GPS

As shown in Figure 66, governments are also investing in quantum computing

companies. For instance, the likes of Innovate UK (part of the UK Research and
Innovation agency) and the National Science Foundation (a U.S. governmental
organization that helps start-ups through grants and seed funding) are also making
a large number of investments.

Figure 66. Most Active Investors in Quantum Computing Companies (1Q 2023)
Investors Investments Investor Type Country
Innovate UK 42 Government United Kingdom
Company A 24 Venture Capital France
European Innovation Council Fund 23 Government Belgium
Creative Destruction Lab 18 Accelerator/Incubator Canada
Company B 17 Venture Capital United States
Company C 11 Venture Capital United States
Oxford Science Enterprises 11 University United Kingdom
National Science Foundation 10 Government United States
Company D 9 Corporate Venture Capital United States
Company E 9 Venture Capital United Kingdom
Company F 9 Venture Capital Germany
Company G 8 Venture Capital Singapore
Company H 8 Venture Capital United States
In-Q-Tel 8 Not-For-Profit Venture Capital United States
Source: PitchBook Inc., Citi GPS

We spoke to David Moehring, General Partner at Cambium Capital, about some of

the challenges quantum computing companies face in raising capital in the current
funding environment.

Expert Interview with David Moehring, General Partner at

Cambium Capital
Q: What are the important characteristics you look for when investing in a
quantum computing company?

David: Quantum computing is a highly complex and multi-faceted technology.

There are many important characteristics, but more important than any one metric
in isolation is the interdependence of that metric to others. Most companies will
strongly highlight one specific area where they excel, but it is critical to understand
where they do not, and to make sure there are no single points of failure. In such a
complex industry, it is far more important to understand a company’s weaknesses
than their advertised strengths. This is true not just with quantum computing, but
with advanced-computing architectures as a whole.

Q: Can you give some examples?

David: From the engineering perspective, some technologies work very well at
small scale, and may even have years of documented success in a university
David Moehring
setting, but many have insurmountable hurdles to scaling. From the business
General Partner, Cambium Capital
perspective, many companies build hardware without a bona fide use-case in mind.
Further, even if the compute infrastructure is matched to a business case, it is also
important to offer a practical compiler and software stack for the solution to reach its
full potential. The mismatch of building hardware that isn’t viable for end-user
applications often creates inefficiencies in the full value chain that completely
negates the effort and cost of building the computer in the first place.

Q: What are the different types of quantum computing companies?

David: While our fund has only invested in companies building quantum computing
hardware, our partnership has firsthand research and development experience in
academia, government, and industry, as well as in building startups and funding all
aspects of the quantum- and advanced-computing ecosystem. As with classical
computing, no one company can accomplish everything, and thus also with
quantum computing it is important to have a robust industry spanning design,
fabrication, manufacturing, software, and application development. Some
companies encompass multiple of these aspects, but regardless, all must be
addressed to unlock the full potential for quantum computing.

How To Prepare: A Deeper Understanding

Understanding quantum computing companies brings its own set of challenges.
With the industry itself arguably just coming out of its own start-up phase, and only
a handful of quantum computing companies in the process of going public, it is hard
for market participants to assess the longer-term opportunity.

One key thing to recognize is that quantum computing itself is highly nuanced. As
such, understanding quantum computing companies requires a genuine
understanding of the technology, the industry use cases, and the landscape at
large.

For instance, when assessing a hardware company, an understanding of each of

the different qubit technologies and the unique challenges it faces in terms of
reaching scale would be beneficial — and this is before any consideration of
whether the management team’s approach to solving it is the best. Similarly, on the
software side, as we touched on earlier in the report, understanding how quantum
computing algorithms work requires at least some understanding of how quantum
hardware and software interact.

Furthermore, while quantum computing is potentially a genuinely transformational

technology, its status as such makes it difficult to separate bold claims from hype.
While there are numerous established companies in the space, there are also many
start-ups, and it is difficult to independently verify their specific actions,
achievements, or the ramifications of their findings. For instance, companies often
compare the results of their quantum computers to those of powerful — but not
necessarily the best — classical computers. Another approach companies take is to
benchmark quantum computers on highly specific tasks that are designed to test
hardware from an academic perspective. While this may be scientifically valid, it
does not necessarily translate into generating commercial value.

Fundamentally, quantum computing is an industry facing high expectations, as

reflected in the high CAGRs predicted by almost all third-party market forecasts we
identified in the “Market Forecasts” section. In addition to some of the deep science
areas mentioned above, numerous quantum computing-related services are
provided via cloud companies, IT service companies, supply chain, and education
providers.

We spoke to Mark Danchak, Partner at General Innovation Capital, about what

market participants should consider when it comes to investigating quantum
computing.

Expert Interview with Mark Danchak, Partner at General

Innovation Capital
Q: What information is important for understanding the quantum computing
investing space?

Mark: I have spent six years building an information matrix that we can cross-
reference to readout who we think the most likely winners will be in the quantum
computing (QC) space. Each underlying substrate that is being explored to build
scaled systems has its own unique advantages and disadvantages, both for building
systems and running algorithms. That level of nuanced understanding is difficult to
see without having spent a great deal of time with these companies. It is not easy
for the traditional venture fund to pick winners in the space.

Q: What are the main vectors of development you are tackling?

Mark: The QC industry will take off when three vectors of progress intersect: (1)
scaled hardware systems, (2) error-correction methodologies, and (3) algorithm
development. If you can get a handle on these, you can start to factor in one of the
Mark Danchak
most important elements of venture investing, which is the question of when
Partner, General Innovation Capital
customers will start to show up meaningfully. I hear people throw out statements
such as “quantum is always five years away,” which is an old trope people have
used for many technologies, but I can tell you 95% of the people looking at this
space today barely heard the words quantum computing four or more years ago.

Q: Which industries do you think will benefit first from quantum computing?

Mark: The industries that we believe will see the early use cases for quantum
computing will be one of two archetypes: (1) companies that deal in the quantum
mechanical realm as a fundamental part of their everyday business, such as
pharmaceuticals, advanced materials, and chemicals; and (2) companies that can
achieve large gains by leveraging small gains in computational areas, such as
(perhaps primarily) finance.

Closing Statement
The newness and complexity of quantum computing is a challenge for stakeholders,
but fostering awareness of what these machines can achieve could help with
preparation and execution. We hope this report shines a brighter light on some of
the opportunities, issues, and timelines around the technology. The debate around
the arrival of quantum advantage will continue, and it will appear at different times
for different use cases. When it does, despite the upskilling and technology
integration that will be necessary, quantum computing is likely to scale
exponentially.

Appendix: How Quantum Computers

Work
We recognize that understanding the physics underpinning quantum computing is
not necessary in order to appreciate the potential impact it may have on the world,
in the same way that users of classical computers do not need to understand how a
transistor works. However, quantum physics is often portrayed as an esoteric and
almost mystical area of science in the media. For that reason, we felt it would be
diligent to provide a basic conceptual grounding for some of the terms that are often
used when describing quantum computing, including the likes of “superposition,”
“entanglement,” and “interference.”

The Basics of Quantum Physics

Quantum physics is, in essence, the study of energy and matter at the smallest of
scales, and its aim is to understand the properties and behavior of the building
blocks of nature. One analogy would be to imagine the world around you as a digital
image made up of trillions of pixels; in such a case, quantum physics would be the
investigation of what happens in between those pixels. We could easily write a book
just on the history of quantum physics, but for the purposes of this report, we will
focus on giving you a brief history and what you need to know from it to get a grasp
on how a quantum computer works.

The field of quantum physics began in the early 1900s following experimental
observations of atoms that seemed to contradict the then classical physics
paradigm. One of the realizations was that energy and matter on the smallest of
scales can be thought of as discrete packets, or “quanta,” each with their own
associated minimum value. In the case of light, energy is delivered by particles of
light known as “photons.” Each photon of a particular frequency delivers the exact
same amount of energy and, most importantly, cannot be broken down into smaller
units. The same applies for an electrical current, in which the “electron” is a
fundamental carrier of charge.

The Two-Slit Experiment

If there is one experiment that can provide a conceptual understanding of quantum
physics, it is the famous two-slit experiment.

The experiment shows that when light waves reach two narrow slits positioned very
close to each other, the light waves interfere with one another and result in what we
refer to as an interference pattern of light and dark areas on the screen behind. This
is because light waves, like all waves, are characterized by peaks and troughs, and
“interference” simply refers to how these light waves combine.

Figure 67. Double-Slit Experiment (with Light Waves)

Source: Shutterstock

Constructive interference is when two light waves meet in-phase — that is to say,
when the peaks of two waves meet one another. This results in the two waves
combining “constructively,” increasing the size of the resultant wave’s peaks and
troughs, and in the case of light, resulting in an area of greater brightness.
Destructive interference is when two light waves meet out-of-phase — that is to say,
when the peak of one wave meets the trough of the other. This results in the two
waves combining “destructively,” cancelling out each other’s peaks and troughs,
and in the case of light, resulting in an area of darkness.

Figure 68. Constructive vs. Destructive Interference

Source: Shutterstock

These patterns of light and darkness reflect areas of constructive and destructive
interference, in the same way you see ripples from two separate waves overlapping
combining to create peaks and troughs in a pond.

If we repeat the experiment with a beam of electrons — something once believed to

be a point particle — we find that we observe the same interference patterns.

Figure 69. Double-Slip Experiment (with Electrons)

Source: Shutterstock

While the natural assumption would be that the particles are colliding with one
another after going through the two slits, even when we fire one electron at a time at
the double-slit, we eventually obtain the same interference patterns. This leads us
to one of the key principles of quantum mechanics: the idea of wave-particle duality,
meaning that on an atomic scale, matter can behave as both a particle and a wave.

The wave-particle duality of nature explains what actually happens — that the
electron behaves as a wave at the point that it reaches the double-slit and, in fact,
takes both paths at the same time and interferes with itself. This notion of an
electron taking multiple paths at the same time is what we know as the principle of
“superposition.” Superposition is a term used to describe an object in a combination
of multiple position states at the same time. This effect contributes to the power of
quantum computing, as it means multiple computations can be carried out in
parallel.

Figure 70. How the Act of Observation Collapses a Superposition

Source: Citi GPS

However, things get weirder still in the world of quantum mechanics. When
attempting to observe this superposition of states of an electron, scientists found
that the electron began to behave as a point particle, with the rear-screen showing
only two distinct areas where the electrons had made contact and without any of the
interference pattens seen before. This leads up to another key concept in quantum
computing, which is that the act of observing a superposition in fact collapses it.
Thus, with respect to quantum computing, though many computations can be
carried out in parallel, you can observe only one of the results at a time. This is why
separate computations have to be combined in a clever way before an observation
is made from a quantum computer, ultimately ensuring that the correct answer has
a high probability of being measured.

The Basics of Computing

How Classical Computers Work
To help understand what makes a quantum computer special, a basic
understanding of how our current “classical” computers work is beneficial. As noted
in our “Understanding Quantum Computers” section, classical computers rely on
bits that are only ever in one state at a time — either on or off. You can imagine it
almost like a light switch, with “on” corresponding to a 1 and “off” corresponding to
0.

Next is to understand the concept of a logic gate, which undertakes simple

mathematical calculations. The easiest to understand is probably the “AND gate,”
which simply outputs a “1” if both the first input AND the second input are equal to 1.
These logic gates are formed from transistors. Consequently, the more transistors
we can place on a computer chip, the more logic gates we can have and the more
calculations a computer can undertake in a given time (i.e., the faster it gets).

Furthermore, modern classical computers have come a very long way in the past 70
years, and part of their progress is attributable to their standardized design. While of
course that period has seen countless innovations in computers’ underlying
architecture, all have used the same exact material — the silicon transistor — to
facilitate the computation.

Figure 71. Logic Gates: Symbols and Truth Tables

Logic Gates – Symbols and Truth Tables

AND NOT NAND

(Inverter) (NOT
AND)
Input 1 Input 1
Output Input Output Output
Input 2 Input 2

Input 1 Input 2 Output Input 1 Output Input 1 Input 2 Output

0 0 0 0 0 1
0 1
0 1 0 0 1 1
1 0 0 1 0 1
1 0
1 1 1 1 1 0

Source: Citi GPS

Quantum computers, on the other hand, come in two different structural models and
use an even greater variety of qubit technologies (see the “The State of the
Quantum Computing Market” section).

Comparing Quantum Computers to Classical Computers

In contrast to the highly standardized way we build classical computers, quantum

computers are created in many different ways. First proposed as a concept by
famed physicist Richard Feynman in the 1980s, the quantum computer relies
instead on a quantum bit, or “qubit.” Unlike classical bits that can only be in either 0
or 1, qubits can also be in a superposition of 0 and 1. Furthermore, each additional
qubit added to a system results in an approximate doubling of its power (for certain
applications).

Since quantum physics is observed across multiple different forms of energy and
matter, there is currently no standardized way of creating a qubit. Figure 72 below
shows the main similarities and differences between classical and quantum
computers.

Figure 72. Classical Hardware vs. Quantum Hardware

Source: Citi GPS

How Qubits Make Quantum Computers So

Powerful
The best way to imagine a qubit is through the conceptual idea of a Bloch sphere,
which is a geometric representation of a qubit and the various states it can occupy.

The best way to imagine a qubit is through the conceptual idea of a Bloch sphere,
which is a geometric representation of a qubit and the various states it can occupy.

Figure 73. Qubits vs. Bits

0 0

1 1
Source: Citi GPS

Just like how a classic bit has two states of either 0 or 1, a qubit has two
computational basis states positioned on opposing poles of the Bloch sphere, as
shown in Figure 73.

These two basis states are the only values returned when measuring the qubit,
analogous to how the electron travels through only one of the two slits in the two-slit
experiment as a particle when being observed (because any observation or
measurement of a quantum state collapses its superposition). The rest of the time,
when the qubit is not observed or measured in any way, it remains in a
superposition of states, which can be visualized as an arrow pointing to a position
on the surface of the sphere other than the north or south poles.

The surface of the sphere represents all possible states that the qubit can be in. In
our example, the arrow points to a specific state. The latitude of the tip of the arrow
indicates the likelihood of measuring a 0 or 1.

Figure 74. Measurement of One Qubit

Source: Citi GPS

Another key difference between classical computers and quantum computers is that
all classical bits are independent of one another, meaning that the state of one bit
does not impact the state of another bit. What makes quantum computers so
powerful are not just the qubits used, but also the ability to “entangle” the qubits.

Qubits display a special phenomenon known as entanglement, whereby individual

qubits can become intrinsically linked and thus no longer act independently of one
another. What this means in practice is that each of the two constituent qubits no
longer has its own precise individual state, indicated by the arrow not reaching the
surface of the sphere in Figure 75showing two qubits in an entangled state, but
instead part of their identity resides in the link between them. This manifests in the
appearance of correlations in the joint probability distribution for the two
measurement outcome bits.

Figure 75. Two Qubits in an Entangled State

Source: Citi GPS

Figure 76. Two Qubits Not in an Entangled State

Source: Citi GPS

And for specific problems, this is one of the keys to the power of quantum
computers — that through the process of entanglement, just by adding one
additional qubit (which, when measured, still only provides either a 0 or 1), you can
actually double the total number of states the quantum computer can be in at a time
and thus approximately double the calculating power for the entire system.

Figure 77. Number of Potential States That Can Simultaneously Exist

Source: Citi GPS

As with any exponential relationship, the higher the number of qubits, the greater
the rate of increase in capabilities (or the steeper the curve, in the case in Figure
77). But to actually harness this power, you need to design a clever algorithm.

Altering the states of individual qubits in this entangled system allows us to alter the
overall state, and thus the probability distribution of the measured outcome.
Ultimately, while a quantum computer can be in a superposition of thousands or
millions of computational basis states at the same time, measuring the qubit state
collapses the superposition to one resultant individual outcome. In addition, one can
use “constructive interference” to increase the likelihood of finding the right answer,
and “destructive interference” to decrease the likelihood of getting the wrong
answer.

Figure 78. How Entangling Qubit Results Generates a Probability Distribution

Source: Citi GPS

It is very important to note that quantum computers, unlike classical computers, do

not provide exact answers, but rather probability distributions. If created well, these
distributions will identify the most probable answer. This can be useful since
quantum computers enable the solving of problems that would otherwise be
intractable on classical computers. The most famous example is Peter Shor's
algorithm, which showed it was possible to break current encryption techniques.
However, not all problems will necessarily be solved faster with a quantum
computer, and there is a field of study known as quantum complexity theory that
addresses this issue specifically — something we discussed in the “A Hybrid
Approach to Computing Is Inevitable” section earlier in the report.

Figure 78 provides only a very high-level overview of how entangling qubits results
in a probability distribution. Rather, the measurement of such an entangled state of
qubits results in both imaginary and real parts — as can be seen in Figure 79 and
Figure 80 below, which show entangled systems of two and three qubits,
respectively.

Figure 79. Probability Distribution Generated from Observations of Two Entangled Qubits

Source: Citi GPS

Figure 80. A Probability Distribution Generated from Observations of Three Entangled Qubits

Source: Citi GPS

© 2023 Citigroup
As our premier thought leadership product, Citi Global Perspectives &
Solutions (Citi GPS) is designed to help readers navigate the most demanding
challenges and greatest opportunities of the 21st century. We access the best
elements of our global conversation with senior Citi professionals, academics, and
corporate leaders to anticipate themes and trends in today’s fast-changing and
interconnected world.

All Citi GPS reports are available on our website www.citi.com/citigps

Future of Nuclear Energy in Voluntary Carbon Market

a Low-Carbon Environment A Critical Piece of the Net
Fission and Fusion Advanced Zero Puzzle
Reactors to Prevail July 2023
July 2023

Economic and Social Sustainable Ocean

Mobility Economy
The Role of Business in Charting a Prosperous Blue
Improving Outcomes Future; Risk to Resilience
June 2023 June 2023

Asia as a Time Machine to Money, Tokens, and Games

the Future Blockchain’s Next Billion
Seven Areas Where Asia Users and Trillions in Value
Gives Insights Into the Future March 2023
May 2023

The Cyber Problem The Creator Economy

Causes and Consequences of Getting Creative and Growing
the Rise in Cyber Skill March 2023
Demand
March 2023

Generative AI Supply Chain Finance

ChatGPT and Search Uncertainty in Global Supply
February 2023 Chains Is Going to Stay
January 2023

State of Global Electric Disruptive Innovations IX

Vehicle Adoption Ten More Things to Stop and
A Trip Around the World Think About
January 2023 December 2022

Antimicrobial Resistance Climate Finance

The Silent Pandemic Mobilizing the Public and
December 2022 Private Sector to Ensure a
Just Energy Transition
November 2022

Food Security Energy Transition: Vol 1

Tackling the Current Crisis Mixed Momentum on the Path
and Building Future to Net Zero
Resilience November 2022
November 2022
Energy Transition: Vol 2 China's Inward Tilt
Building Bridges to Renew The Pursuit of Economic Self-
Momentum Reliance
November 2022 October 2022

Philanthropy v2.0 Food and Climate Change

Reinventing Giving in Sustainable Foods Systems
Challenging Times for a Net-Zero Future
October 2022 July 2022

Home of the Future 2 Global Supply Chains

PropTech – Towards a The Complexities Multiply
Frictionless Housing Market? June 2022
June 2022

Space Investing for Outcomes

The Dawn of a New Age Why Impact Is Relevant
May 2022 Beyond Impact Investing
April 2022

Metaverse and Money Global Art Market

Decrypting the Future Disruptions
March 2022 Pushing Boundaries
March 2022

Women Entrepreneurs Eliminating Poverty

Catalyzing Growth, The Importance of a
Innovation, and Equity Multidimensional Approach
March 2022 February 2022

Global Supply Chains Philanthropy and the Global

The Complicated Road Back Economy
to “Normal” Opportunities in a World of
December 2021 Transition
November 2021

Education: Learning for Life Home of the Future

Why L&D Is the Next Frontier Building for Net Zero
in Global Education
November 2021 October 2021

Global Carbon Markets Disruptive Innovations VIII

Solving the Emissions Crisis Ten More Things to Stop and
Before Time Runs Out Think About
October 2021 October 2021

Holistic Digital Policy Biodiversity

Nation States Must Lead in The Ecosystem at the Heart of
Building Equitable Human- Business
Centric Digital Economies July 2021
October 2021
July 2023 Citi GPS: Global Perspectives & Solutions 147

If you are visually impaired and would like to speak to a Citi representative regarding the details of the graphics in this
document, please call USA 1-888-800-5008 (TTY: 711), from outside the US +1-210-677-3788

IMPORTANT DISCLOSURES
This communication has been prepared by Citigroup Global Markets Inc. and is distributed by or through its locally authorised affiliates (collectively, the "Firm")
[E6GYB6412478]. This communication is not intended to constitute "research" as that term is defined by applicable regulations. Unless otherwise indicated, any reference to a
research report or research recommendation is not intended to represent the whole report and is not in itself considered a recommendation or research report. The views
expressed by each author herein are his/ her personal views and do not necessarily reflect the views of his/ her employer or any affiliated entity or the other authors, may differ
from the views of other personnel at such entities, and may change without notice.
You should assume the following: The Firm may be the issuer of, or may trade as principal in, the financial instruments referred to in this communication or other related
financial instruments. The author of this communication may have discussed the information contained herein with others within the Firm and the author and such other Firm
personnel may have already acted on the basis of this information (including by trading for the Firm's proprietary accounts or communicating the information contained herein to
other customers of the Firm). The Firm performs or seeks to perform investment banking and other services for the issuer of any such financial instruments. The Firm, the Firm's
personnel (including those with whom the author may have consulted in the preparation of this communication), and other customers of the Firm may be long or short the
financial instruments referred to herein, may have acquired such positions at prices and market conditions that are no longer available, and may have interests different or
adverse to your interests.
This communication is provided for information and discussion purposes only. It does not constitute an offer or solicitation to purchase or sell any financial instruments. The
information contained in this communication is based on generally available information and, although obtained from sources believed by the Firm to be reliable, its accuracy
and completeness is not guaranteed. Certain personnel or business areas of the Firm may have access to or have acquired material non-public information that may have an
impact (positive or negative) on the information contained herein, but that is not available to or known by the author of this communication.
The Firm shall have no liability to the user or to third parties, for the quality, accuracy, timeliness, continued availability or completeness of the data nor for any special, direct,
indirect, incidental or consequential loss or damage which may be sustained because of the use of the information in this communication or otherwise arising in connection with
this communication, provided that this exclusion of liability shall not exclude or limit any liability under any law or regulation applicable to the Firm that may not be excluded or
restricted.
The provision of information is not based on your individual circumstances and should not be relied upon as an assessment of suitability for you of a particular product or
transaction. Even if we possess information as to your objectives in relation to any transaction, series of transactions or trading strategy, this will not be deemed sufficient for
any assessment of suitability for you of any transaction, series of transactions or trading strategy.
The Firm is not acting as your advisor, fiduciary or agent and is not managing your account. The information herein does not constitute investment advice and the Firm makes
no recommendation as to the suitability of any of the products or transactions mentioned. Any trading or investment decisions you take are in reliance on your own analysis and
judgment and/or that of your advisors and not in reliance on us. Therefore, prior to entering into any transaction, you should determine, without reliance on the Firm, the
economic risks or merits, as well as the legal, tax and accounting characteristics and consequences of the transaction and that you are able to assume these risks.
Financial instruments denominated in a foreign currency are subject to exchange rate fluctuations, which may have an adverse effect on the price or value of an investment in
such products. Investments in financial instruments carry significant risk, including the possible loss of the principal amount invested. Investors should obtain advice from their
own tax, financial, legal and other advisors, and only make investment decisions on the basis of the investor's own objectives, experience and resources.
This communication is not intended to forecast or predict future events. Past performance is not a guarantee or indication of future results. Any prices provided herein (other
than those that are identified as being historical) are indicative only and do not represent firm quotes as to either price or size. You should contact your local representative
directly if you are interested in buying or selling any financial instrument, or pursuing any trading strategy, mentioned herein. No liability is accepted by the Firm for any loss
(whether direct, indirect or consequential) that may arise from any use of the information contained herein or derived herefrom.
Although the Firm is affiliated with Citibank, N.A. (together with its subsidiaries and branches worldwide, "Citibank"), you should be aware that none of the other financial
instruments mentioned in this communication (unless expressly stated otherwise) are (i) insured by the Federal Deposit Insurance Corporation or any other governmental
authority, or (ii) deposits or other obligations of, or guaranteed by, Citibank or any other insured depository institution. This communication contains data compilations, writings
and information that are proprietary to the Firm and protected under copyright and other intellectual property laws, and may not be redistributed or otherwise transmitted by you
to any other person for any purpose.
IRS Circular 230 Disclosure: Citi and its employees are not in the business of providing, and do not provide, tax or legal advice to any taxpayer outside of Citi. Any statements
in this Communication to tax matters were not intended or written to be used, and cannot be used or relied upon, by any taxpayer for the purpose of avoiding tax penalties. Any
such taxpayer should seek advice based on the taxpayer’s particular circumstances from an independent tax advisor.
© 2023 Citigroup Global Markets Inc. Member SIPC. All rights reserved. Citi and Citi and Arc Design are trademarks and service marks of Citigroup Inc. or its affiliates and are
used and registered throughout the world.

NOW / NEXT
Key Insights regarding the future of Quantum Computing

INNOVATION There is currently no known efficient algorithm that can run on a classical
computing to decrypt public-key cryptographic standards. / Through Shor’s
algorithm and Grover’s algorithm, advanced quantum computers can attack
symmetric encryption standards and in the next 15 years, could break 2048-bit RSA
encryption.

LABOR MARKET With only a few thousand quantum scientists and engineers estimated worldwide in
2020, lack of talent could be a hindrance to the technology’s growth. / Short course
and professional development courses are emerging to allow the transfer of skills to
the quantum computing industry.

TECHNOLOGY High R&D costs, high failure rates, and long development cycles are commonplace in
the healthcare industry. / Quantum computers, with their expected future
advantages in optimization, machine learning, and molecular simulation, will provide
the computational toolkit required to some of these entrenched problems.

Artificial Intelligence Valuation
No ratings yet
Artificial Intelligence Valuation
710 pages
Fundamentals of Quantum Computing (2021) (9783030636890) (2021)
100% (6)
Fundamentals of Quantum Computing (2021) (9783030636890) (2021)
480 pages
Quantum Technology Monitor April 2024
No ratings yet
Quantum Technology Monitor April 2024
103 pages
303012357X Mathematics of Quantum Computing An Introduction (Scherer 2019-11-13) (29B45CBD) PDF
100% (5)
303012357X Mathematics of Quantum Computing An Introduction (Scherer 2019-11-13) (29B45CBD) PDF
773 pages
GDC EMEA Whitepaper Quantum-Computing EN
No ratings yet
GDC EMEA Whitepaper Quantum-Computing EN
17 pages
Quantum Computing, 2nd Ed.
100% (4)
Quantum Computing, 2nd Ed.
223 pages
Quantum Technologies Global Market - Forecast To 2030
No ratings yet
Quantum Technologies Global Market - Forecast To 2030
6 pages
Concise Guide To Quantum Computing: Sergei Kurgalin Sergei Borzunov
75% (4)
Concise Guide To Quantum Computing: Sergei Kurgalin Sergei Borzunov
129 pages
Cessna Citation CE-550 JT-15: Note Taking Guide
100% (1)
Cessna Citation CE-550 JT-15: Note Taking Guide
20 pages
(IMF Working Papers) Quantum Computing and The Financial System - Spooky Action at A Distance
No ratings yet
(IMF Working Papers) Quantum Computing and The Financial System - Spooky Action at A Distance
14 pages
Quantum Computing A New Frontier in Techno-Nationalism - Hinrich Foundation - Alex Capri - August 2021
No ratings yet
Quantum Computing A New Frontier in Techno-Nationalism - Hinrich Foundation - Alex Capri - August 2021
27 pages
BCG The Next Decade in Quantum Computing Nov 2018 21 R Tcm9 207859
No ratings yet
BCG The Next Decade in Quantum Computing Nov 2018 21 R Tcm9 207859
30 pages
Quantum Technology
No ratings yet
Quantum Technology
26 pages
Frontiers in Quantum Computing - New Research (2022) (Anandan) (9781685078164) (2022)
No ratings yet
Frontiers in Quantum Computing - New Research (2022) (Anandan) (9781685078164) (2022)
280 pages
Accenture Get Ready For The Quantum Impact
No ratings yet
Accenture Get Ready For The Quantum Impact
25 pages
Oida 2020 3
No ratings yet
Oida 2020 3
69 pages
Quantum Computing in Banking - Indian
No ratings yet
Quantum Computing in Banking - Indian
5 pages
Quantum Computing
100% (2)
Quantum Computing
27 pages
Innovating With Quantum Computing: Enterprise Experimentation Provides View Into Future of Computing
100% (1)
Innovating With Quantum Computing: Enterprise Experimentation Provides View Into Future of Computing
16 pages
Quantum Computing For Finance State-of-the-Art and
No ratings yet
Quantum Computing For Finance State-of-the-Art and
24 pages
Quantum Computing
100% (4)
Quantum Computing
31 pages
Hughes2021 Book QuantumComputingForTheQuantumC
100% (6)
Hughes2021 Book QuantumComputingForTheQuantumC
159 pages
Quantum Computing
No ratings yet
Quantum Computing
14 pages
Quantum Computing and Finance
No ratings yet
Quantum Computing and Finance
25 pages
Quantum Technologies Execsum
No ratings yet
Quantum Technologies Execsum
24 pages
Global Future Council On Quantum Computing Frequently Asked Questions
No ratings yet
Global Future Council On Quantum Computing Frequently Asked Questions
4 pages
Quantum Algo Impl Beginners 1804.03719
100% (1)
Quantum Algo Impl Beginners 1804.03719
96 pages
Quantum Computing 101. Understanding The Basics 2023
100% (2)
Quantum Computing 101. Understanding The Basics 2023
31 pages
Gokul Sai R:-1GA17EC041 Harshitha H R: - 1GA17EC049 Gowrav S: - 1GA17EC042 Amogh R: - 1GA17EC013 Bhoomika S S:-1GA17EC019
100% (1)
Gokul Sai R:-1GA17EC041 Harshitha H R: - 1GA17EC049 Gowrav S: - 1GA17EC042 Amogh R: - 1GA17EC013 Bhoomika S S:-1GA17EC019
14 pages
Recent Development in Integrated Lithium Niobate Photonics
No ratings yet
Recent Development in Integrated Lithium Niobate Photonics
45 pages
Quantum Programming in QCL PDF
No ratings yet
Quantum Programming in QCL PDF
109 pages
Quantum Technologies Roadmap Report Univ Cambrige UK
No ratings yet
Quantum Technologies Roadmap Report Univ Cambrige UK
76 pages
The Neural Qubit
No ratings yet
The Neural Qubit
10 pages
Quantum-Computing Accenture
100% (1)
Quantum-Computing Accenture
20 pages
Principles of Superconducting Quantum Computers
No ratings yet
Principles of Superconducting Quantum Computers
380 pages
Adiabatic Computing
No ratings yet
Adiabatic Computing
25 pages
IDQ White Paper - IDQ On Quantum Technologies
No ratings yet
IDQ White Paper - IDQ On Quantum Technologies
9 pages
Goldman Sachs Book List
No ratings yet
Goldman Sachs Book List
4 pages
Azure Quantum - An Emerging Quantum Computing Ecosystem
No ratings yet
Azure Quantum - An Emerging Quantum Computing Ecosystem
8 pages
Quantum Computing.8166951.Powerpoint
No ratings yet
Quantum Computing.8166951.Powerpoint
7 pages
Gs 16 18 Quantum Technologies Report
100% (1)
Gs 16 18 Quantum Technologies Report
64 pages
FSO System Modules and Design Issues: January 2017
No ratings yet
FSO System Modules and Design Issues: January 2017
29 pages
303019065X A Primer On Quantum Computing (De Lima Marquezino, Portugal & Lavor 2019-06-27) (D48754E7) PDF
100% (3)
303019065X A Primer On Quantum Computing (De Lima Marquezino, Portugal & Lavor 2019-06-27) (D48754E7) PDF
114 pages
Instant Download Supervised Learning With Quantum Computers Maria Schuld PDF All Chapters
100% (3)
Instant Download Supervised Learning With Quantum Computers Maria Schuld PDF All Chapters
62 pages
Quantum Mechanics - Engineering Physics
No ratings yet
Quantum Mechanics - Engineering Physics
17 pages
Quantum Neural Network Classifiers: A Tutorial: Submission
No ratings yet
Quantum Neural Network Classifiers: A Tutorial: Submission
30 pages
Download full Relativity Gravitation and Cosmology A Basic Introduction 2nd Edition Ta-Pei Cheng ebook all chapters
50% (2)
Download full Relativity Gravitation and Cosmology A Basic Introduction 2nd Edition Ta-Pei Cheng ebook all chapters
45 pages
Quantum Computing
No ratings yet
Quantum Computing
33 pages
Progress and Visions in Quantum Theory in View of Gravity
No ratings yet
Progress and Visions in Quantum Theory in View of Gravity
302 pages
Amira Abbas: Quantum Computing Lecture at Wits University PDF
No ratings yet
Amira Abbas: Quantum Computing Lecture at Wits University PDF
102 pages
Quantum Computing Lectures Preskill
100% (4)
Quantum Computing Lectures Preskill
321 pages
Bricmont, J. (2017) - Quantum Sense and Nonsense. Basel Springer (c31)
No ratings yet
Bricmont, J. (2017) - Quantum Sense and Nonsense. Basel Springer (c31)
287 pages
Introduction To Quantum Technologies
No ratings yet
Introduction To Quantum Technologies
298 pages
Quantum Market Research
No ratings yet
Quantum Market Research
56 pages
Post Quantum Cryptography Michael Hamburg Hardwear Io Usa 2019
No ratings yet
Post Quantum Cryptography Michael Hamburg Hardwear Io Usa 2019
88 pages
Full Download Quantum Computing for Everyone (The MIT Press) Chris Bernhardt PDF DOCX
100% (1)
Full Download Quantum Computing for Everyone (The MIT Press) Chris Bernhardt PDF DOCX
40 pages
WEF量子计算的现状：构建量子经济 2022
No ratings yet
WEF量子计算的现状：构建量子经济 2022
48 pages
Quantum computing primer
No ratings yet
Quantum computing primer
35 pages
ADL Myths Applications Impacts QC 2023
No ratings yet
ADL Myths Applications Impacts QC 2023
12 pages
Disrupting Traditional Industries With Quantum Computing
No ratings yet
Disrupting Traditional Industries With Quantum Computing
6 pages
Quantum Technology a Note-1-3
No ratings yet
Quantum Technology a Note-1-3
3 pages
History of Battery Development
No ratings yet
History of Battery Development
24 pages
Mango Leaf CNN
No ratings yet
Mango Leaf CNN
26 pages
Doctorate Programmes - Faculty of Built Environment and Surveying
No ratings yet
Doctorate Programmes - Faculty of Built Environment and Surveying
6 pages
Guidelines and Instructions To Candidates
No ratings yet
Guidelines and Instructions To Candidates
3 pages
"Royal Industrial Estate": Project Report Brief Details of The Firm
No ratings yet
"Royal Industrial Estate": Project Report Brief Details of The Firm
5 pages
Resume Final
No ratings yet
Resume Final
4 pages
L6 All Together
No ratings yet
L6 All Together
68 pages
Entrepreneurship Development Course Code: 22032: Final Dt. 1
No ratings yet
Entrepreneurship Development Course Code: 22032: Final Dt. 1
31 pages
Name: Toheeb Adewale Tijani Wallet Number: 7063049182
No ratings yet
Name: Toheeb Adewale Tijani Wallet Number: 7063049182
14 pages
Technical Brief: PVC Pressure Pipe Under Roads
No ratings yet
Technical Brief: PVC Pressure Pipe Under Roads
1 page
Answer Key - CK-12 Chapter 04 Basic Algebra Concepts
No ratings yet
Answer Key - CK-12 Chapter 04 Basic Algebra Concepts
29 pages
CASE STUDY - IOCL HR - Furnace Incident - ED Approved
No ratings yet
CASE STUDY - IOCL HR - Furnace Incident - ED Approved
3 pages
Galaxy VX: Installation
No ratings yet
Galaxy VX: Installation
92 pages
J2.0-3.5XTLG (Outdoor Electric Forklift (A2A6 - Hyster J2.0-3.5XLTG - Brochure)
No ratings yet
J2.0-3.5XTLG (Outdoor Electric Forklift (A2A6 - Hyster J2.0-3.5XLTG - Brochure)
8 pages
Sub Grade
No ratings yet
Sub Grade
6 pages
7 - Configuring A VPLS
No ratings yet
7 - Configuring A VPLS
9 pages
About BSNL Live:: Video On Demand
No ratings yet
About BSNL Live:: Video On Demand
13 pages
Proteus Pro 8 Sp3 Full 17 PDF
No ratings yet
Proteus Pro 8 Sp3 Full 17 PDF
3 pages
Samsung ProRo4u TV Service Manual
No ratings yet
Samsung ProRo4u TV Service Manual
174 pages
Dell Latitude E6410 Compal LA-5472P Discrete Schematics
No ratings yet
Dell Latitude E6410 Compal LA-5472P Discrete Schematics
66 pages
Foundations of Artificial Intelligence
No ratings yet
Foundations of Artificial Intelligence
60 pages
2.2 T4594-G-QTT-002 - HURE - Kouroussa LV Substations Siemens Solution BOQ
No ratings yet
2.2 T4594-G-QTT-002 - HURE - Kouroussa LV Substations Siemens Solution BOQ
82 pages
Lista D
No ratings yet
Lista D
5 pages
CSEC Add Maths 2022 June Past Paper Solutions
No ratings yet
CSEC Add Maths 2022 June Past Paper Solutions
27 pages
My Capital One-1 (2) - 1
No ratings yet
My Capital One-1 (2) - 1
1 page
Polymath Tutorial On Non-Linear Regression (Example 7-4) : A Pco 2 1+ 2
No ratings yet
Polymath Tutorial On Non-Linear Regression (Example 7-4) : A Pco 2 1+ 2
10 pages
EE8552 POWER ELECTRONICS Syllabus
No ratings yet
EE8552 POWER ELECTRONICS Syllabus
1 page
Operations Research 289706179
No ratings yet
Operations Research 289706179
39 pages
Enquiry Letters Assigment Wahyu Jonathan Zakky
No ratings yet
Enquiry Letters Assigment Wahyu Jonathan Zakky
5 pages