1 INTRODUCTION TO AI

EXCITE
Information Technology and Artificial Intelligence
The computer which we see today was not like this before. From abacus to modern computers,
technology has come a long way. This fascina ng journey is very old. Let us first look at the
history and genera ons of the computer.
Abacus, one of the earliest known calcula ng devices, was used
in many great ancient civiliza ons and had sliding counters
(beads, pebbles or flat disks) to perform calcula ons. The first
computer which laid the founda on for today's modern
Abacus
machines was the Analy cal Engine, a device conceived and
designed by the Bri sh mathema cian Charles Babbage
between 1833 and 1871. Another milestone in computer
science was achieved when the electromechanical computer Analytical Engine
Mark 1 was built in the early 1940s. It used punched cards or
punched tape for its program input. In the early stages of
compu ng, most computer opera ons were simple calcula ons.
In 1936, Alan Turing introduced the idea of a universal Mark I
machine, later named the Turing machine, capable of
compu ng anything that is computable. The key principles of
the modern computer was based on his ideas. By late 1960s,
computer systems included screens, keyboards, text editors, and
used programming languages such as FORTRAN and COBOL. Turing Machine
The Internet was first created in the 1970s, but its exposure
was only limited to researchers in universi es for more than a
decade. It was the inven on of the World Wide Web, which
began in 1990 that transformed the Internet into an accessible Modern Computer
vehicle for informa on sharing.
The Internet has transformed and revolu onized our world. It
has made it possible for all of us to communicate quickly and
has brought informa on from all over the world to our
finger ps. This has brought about a revolu on in Informa on
Technology.
Internet
The large amount of data on the Internet and progress in
computer hardware as well as programming techniques is giving
rise to very exci ng new fields like Ar ficial Intelligence.

Artificial Intelligence

PM Publishers Pvt. Ltd.

11
Evolution of Artificial Intelligence
In the past few years, AI has evolved into a powerful tool that enables machines to think and
act like humans. The term ‘Ar ficial Intelligence’ was coined by John McCarthy at an academic
conference he organized in Dartmouth College in 1956 to discuss the subject. He is widely
considered as the father of Ar ficial Intelligence for his pioneer role within the field. The
research in AI was ini ally restricted to problem solving and symbolic methods during the
1950s. Let us learn more about the meline of innova ons in AI using the table given below:

Year Innova on
1950 Ar ficial Intelligence was first conceptualized by Alan Turing to evaluate the
intelligence level of machines. He published a paper en tled Compu ng
Machinery and Intelligence which posed the ques on: Can machines think? He
developed the Imita on Game, popularly known today as the Turing Test, to
determine whether a machine can demonstrate human intelligence.
1956 John McCarthy coined the term ‘Ar ficial Intelligence’ at Dartmouth Conference.

1958 A programming language named LISP was developed for AI by John McCarthy in
MIT.
1964 A disserta on at MIT took place which showed that computers can understand
natural language well enough to solve algebra word problems correctly.
1969 Shakey, a robot, equipped with locomo on, percep on, and problem solving was
built at Stanford Research Ins tute.
1973 A Sco sh robot named Freddy was built at Edinburgh University which was
capable of using vision to locate and assemble models.
1975 An autonomous system named MYCIN was built to diagnose bacterial infec ons
and recommend appropriate an bio cs.
1979 Stanford Cart, the first computer-controlled autonomous vehicle was built.
1986 LMI Lambda, a LISP worksta on was introduced.

1997 Chess playing computer Deep Blue by IBM won against chess world champion
Garry Kasparov.
2003 Computer Science and Ar ficial Intelligence Laboratory (CSAIL) at MIT was formed.
2011 Apple introduced a personal voice assistant Siri as a built-in feature for iOS.
Supercomputer IBM Watson won against two human contestants in the television
quiz show Jeopardy!.
2012 Google’s autonomous car passed the first US self-driving car test.
2014 Amazon launched Alexa – an intelligent virtual assistant.
2017 Google’s AI AlphaGo beat world champion Ke Jie in the board game of Go.
2018 Erica, a robot became a news anchor in Japan.
2019 A gender-neutral voice assistant named Q was created.
2020 Evolu on of chatbots to conversa onal AI bots.

Artificial Intelligence - 9
12
 Artificial Intelligence
Ar ficial Intelligence is a branch of Computer Science which studies how the human brain
thinks, learns, decides and works when it tries to solve problems. Finally, this study helps to
build intelligent so ware systems.
Before moving further in our study of Ar ficial Intelligence, let us explore the concept and
nature of intelligence. As you know, intelligence is intangible. It is composed of:
i) Reasoning: It is the set of processes that enable us to provide a basis for making
judgements, decisions and predic ons.
ii) Learning: It is the ac vity of gaining knowledge or skill by studying, prac cing, being taught
or experiencing something. Learning enhances the awareness of the subjects of study.
iii) Problem Solving: It is the process in which one perceives the problem and tries to arrive at
the desired solu on for the present situa on by analyzing the available alterna ves and
choosing the best path. It also includes decision making.
iv) Percep on: It is the process of acquiring, interpre ng, selec ng and organizing sensory
informa on. In humans, percep on is aided by sensory organs.
v) Linguis c Intelligence: It is one’s ability to use, comprehend, speak and write verbal and
wri en language. It is important in interpersonal communica on.
In our daily life, we perform many tasks, out of which some are regular and some are special.
The performance of regular tasks depends on common intelligence whereas the performance
of special tasks requires special intelligence.
The different types of regular tasks and special tasks for human and computer are given in
the following table:

Category Regular Tasks Special Tasks

Human A human being can rou nely perform these Only a human being with master skills can
tasks. e.g. communica ng in different do special tasks. For example: An eye
languages, moving from one place to another, surgeon can do the surgery of eyes, a
overcoming hurdles, and the ability to think subject teacher can teach a specific subject
conceptually and logically. In simple words, in depth, etc. Thus, we can say that special
we can say that regular tasks performed by skills or special intelligence are required to
human beings do not need specific skills. perform special tasks.
Computer Computer is a smart machine but regular Special tasks which are very easy for
tasks performed by humans are very difficult computers, are difficult for human beings.
for computers. For example: Understanding For example: Check-up of eyesight, medical
any regional language, moving from one diagnosis of illness and doing complex
place to another, and iden fying objects. calcula ons within seconds which is
difficult for human beings.

Tasks which are Hard for Humans but Easy for Computers
A computer can easily solve any complex mathema cal calcula on within a second, which is
extremely difficult or rather impossible for human beings. Let us understand with the help of a
simple example. In a game of chess, humans play chess like a conversa on. For each move, the
player makes a plan whereas the computer doesn't make a plan. It evaluates each posi on
without focusing on the history of the game. Due to this reason, computers started bea ng
chess world champions with the famous Deep Blue game in 1997.

PM Publishers Pvt. Ltd.

13
AI – Combination of Different Disciplines
Ar ficial Intelligence refers to the ability of machines or Computer
Science
computer programs to perform certain ac ons that typically
require human beings to use their intelligence. In simple terms, Engineering Biology
AI is an approach to make a computer, a robot, or a product
to think, act and behave like a human. ARTIFICIAL
INTELLIGENCE
The field of Ar ficial Intelligence is a combina on of different
disciplines like Computer Science, Biology, Psychology, Mathematics Psychology
Philosophy, Mathema cs and Engineering for crea ng
intelligent systems that behave like a human in terms of Philosophy
thinking, learning and differen a ng between right and wrong.

Goals of Artificial Intelligence

The main goal of Ar ficial Intelligence is to aid human
capabili es and help us make advanced decisions with far-
reaching consequences. From a philosophical perspec ve,
Ar ficial Intelligence technology has the poten al to help
humans live more meaningful lives, free from hard labour. It can
help to manage the complex web of interconnected individuals,
companies, states and na ons to func on in a manner that is
universally beneficial to the humanity.
Currently, the purpose of Ar ficial Intelligence is shared by all the different tools and techniques
that we have invented over the past many years – to simplify our efforts and to help us make
be er decisions.
The key goal of Ar ficial Intelligence is to build an intelligent system with the following traits
and capabili es:
1. Reasoning and Problem Solving: AI researchers aim to build systems that strongly resemble
the step-by-step approach of problem solving and logical deduc on shown by humans while
dealing with complex problems.
2. Knowledge Representa on: Knowledge representa on refers to the configura on of
informa on in order to store, retrieve or manipulate it. Since an AI-powered system is
expected to solve diverse problems, it should have extensive knowledge about the world.
3. Planning: The system should have the insight to set future goals and plan accordingly in
advance to achieve them. It should have the capability to visualize future scenarios and
predict future possibili es based on the user’s ac ons.
4. Learning: The AI system should automa cally improve through experience, also known as
machine learning. It should be learning independently without any human interven on.
5. Social Intelligence: AI aspires to develop systems that can recognize, interpret, process,
mimic and predict human behaviour and emo ons.
6. Crea vity: AI system should display out-of-the-box approach to problems and generate
novel, never-seen-before outputs.
7. General Intelligence: Many researchers are trying to integrate Ar ficial General Intelligence
in their systems which will be able to outdo the intelligence and skills of human beings in
every aspect.

Artificial Intelligence - 9
14
 ICE-BREAKER ACTIVITY
BIOMETRIC SYSTEM
Biometrics are physical or behavioral human characteris cs that can be used
to digitally iden fy a person to grant access to systems, devices or data. The
input data of these biometric iden fiers are fingerprints, facial pa erns,
voice or typing cadence. Each of these iden fiers is considered unique to
the individual, and they may be used in combina on to ensure greater
accuracy of iden fica on. Biometrics can provide a reasonable level of
confidence in authen ca ng a person with less fric on for the user, thus it
is more secure.
Now, write any three privacy risks of using biometric systems. How many types of biometrics
are available in India ll now? Can you list some of the technologies used in biometrics?
How secure is biometric authen ca on data?

DO YOU
KNOW ?
By far, the most accurate and convenient form of biometric is fingerprint
iden fica on. Iris recogni on ranks second for security and accuracy. Facial
recogni on leads along with fingerprint iden fica on in terms of convenience.

Artificial Intelligence Vs Human Intelligence

The intelligence level of human beings has come from
past experiences and ac ons based upon different
situa ons and environments. In other words, we can say
that human intelligence is completely based upon the
ability to change one’s surroundings through knowledge
gained from experience. On the other hand, Ar ficial
Intelligence is a way of making intelligent computer
programs or making intelligent smart machines, in a
manner similar to how an intelligent human thinks. A lot of differences exist between human
intelligence and Ar ficial Intelligence on the basis of nature, decision-making, performance,
perfec on, etc. Some of them are as follows:
1. Cogni ve Behaviour: Human intelligence is based on cogni ve behaviour whereas Ar ficial
Intelligence uses models to mimic human behaviour.
2. Nature: Human intelligence is natural whereas Ar ficial Intelligence is ar ficial in nature.
3. Adaptability: Human insights can be adaptable in reac on to the changes to their
environment whereas AI takes much more me to adjust to unfamiliar changes.
4. Decision Making: Human decision making can be biased whereas the decision making
capability of AI is unbiased.
5. Speed: As you know, computers can handle more data at a speedier rate. So, Ar ficial
Intelligence powered machines can solve any complex problem within a millisecond which
human intelligence can never do. In simple words, human intelligence cannot beat the
speed of Ar ficial Intelligence machines.
PM Publishers Pvt. Ltd.
15
Classification of Artificial Intelligence
As defined above, Ar ficial Intelligence refers to the process of simula ng human intelligence in
man-made machines like computer systems. These processes include learning, reasoning, and
self-correc on. In general, Ar ficial Intelligence can be divided into various types. On the basis
of capabili es and func onali es, there are mainly two types:
i) Based on capabili es: Ar ficial Narrow Intelligence (ANI), Ar ficial General Intelligence (AGI)
and Ar ficial Super Intelligence (ASI)
ii) Based on func onali es: Reac ve Machines, Limited Memory, Theory of Mind and Self-
awareness
Type I: Based on Capabilities
When we are classifying AI in this category, it is all about finding a solid ground of skills or
competency for the AI-based machines. These are capability-based AI types:
i) Ar ficial Narrow Intelligence (ANI): Ar ficial Narrow Intelligence is the
ability of computers to perform a single task effec vely and efficiently.
These systems are intelligent enough to perform a single task very well
like Google voice assistants, bots, etc, which are designed only for voice
recogni on. The rapidly growing field of bots serves as an excellent
example of ANI at the workplace. Bots are used to provide immediate
answers to ques ons like “What type of weather is going to be today?” On the basis of the
ques on, bots pull data from larger systems and deliver the expected answer.
Companies like Google and Facebook are striving to create user-friendly bots for a wide variety
of purposes. Let us understand with the help of a simple example of an online departmental
store. Sunita, a customer, wants to order the same product again which she had purchased last
me, but she doesn't remember the exact name of the product. To do this, she would have had
to find the receipt of her last purchase in the previous mes. But, nowadays, it is easier by
logging onto the website and searching the product name through her order history. Using bot,
she could text on a self-service web chat like “What is the name of the product I ordered last
me?” The bot can quickly match her iden ty and find the order history to tell the exact name
of the product.
ii) Ar ficial General Intelligence (AGI): Ar ficial General Intelligence can be defined as the
ability of systems to perform any intellectual task with efficiency
like a human. These systems are hard to develop because
replica on of the human brain is theore cally possible but not
prac cable as of now. Currently, there is no such system which
could come under general AI and can perform any task as
perfectly as a human. So, we can say that Ar ficial General
Intelligence is s ll at the research stage.
iii) Ar ficial Super Intelligence (ASI): Ar ficial Super Intelligence
refers to the ability of a system that could perform any task be er
than humans with cogni ve behaviour. With ASI, a system would
become so self-aware that it would surpass the level of cogni ve
performance, emo onal understanding, problem solving and
decision making skills found in humans. This system is
theore cally possible as it is an outcome of Ar ficial General
Intelligence.
Artificial Intelligence - 9
16
 Type II: Based on Functionalities
As the name suggests, the primary objec ve of such AI-powered systems is to perform some
specific func ons. There are four func onality-based AI types:
i) Reac ve Machines: Reac ve machines, one of the basic forms of AI,
do not have any past memory. These machines cannot use past
experience to take any ac on in future. For example, IBM’s Deep Blue
System, Google’s AlphaGo, etc.
ii) Limited Memory: Limited Memory can be defined as the ability of
machines to store past experiences for a shorter period of me. For
example, driverless cars, Apple’s chatbot Siri, etc.
iii) Theory of Mind: Theory of Mind AI can be defined as the ability of machines to understand
human emo ons, beliefs, thoughts, expecta ons and be able to interact socially like human
beings. Even though they have undergone a lot of improvements, this type of AI machines
are s ll at the research stage.
iv) Self-awareness: Self-aware AI refers to the ability of those machines that are super
intelligent and have their own consciousness, self-awareness and sen ments. In simple
terms, they are essen ally a complete human being. In reality, these types of machines do
not exist. It is only a concept.

Is there Another Way to Classify AI?

Some mes, to differen ate between the performance levels of different Ar ficial Intelligence
systems, scien sts divide them into two types:
a) Weak AI: In the present scenario, all AI systems are weak AI because these systems or
machines pretend to be intelligent but are not really intelligent. Some examples of Weak AI
are advanced chess so ware programs, virtual assistants and language translators. The output
given by these systems are a part of programming which is being done by human beings.
b) Strong AI: Strong AI systems are those systems which use clustering and associa on
techniques for processing the data. The output given by these systems have a wider scope
and applica on and will not be limited to programming. These systems have an incredible
human-level intelligence.

AI Technique
AI technique is defined as a proper organiza on of knowledge to allow its efficient usage. In
this digital age, we have a huge volume of knowledge and data available to us around the
clock. However, most of this data lacks proper forma ng and organiza on, and is prone to a
lot of changes. AI technique helps in proper organiza on, management and forma ng of this
data to make it useful. AI technique helps us to organize this unorganized knowledge by the
following methods:
a) By making it easily modifiable to allow rec fica on of mistakes easier.
b) By making the knowledge easily understood by the person it is being provided to.
c) By making the knowledge useful even if it is incomplete or contains errors.

PM Publishers Pvt. Ltd.

17
 ICE-BREAKER ACTIVITY
DESIGN YOUR DREAM SMART HOME
In the modern era, designing a smart home is everyone’s dream. Proper planning is crucial
to design a smart home that is mindful of the budget, personal style and requirements of its
future inhabitants.
You can follow the given points to design a good smart home:
a) Style: House style plays an important role while designing a smart home. Nowadays,
people want unique house styles customized as per their requirement.
b) Essen als: The basic parts in a typical house are bedroom, kitchen, living area, common
room, dining room, porch, etc. Thus, you should create a list of basic rooms that you
want to have in your smart home.
c) Interior: As you know, a good interior makes your home a rac ve and impressive. While
designing a smart home, you should pay special a en on to the ligh ng system, color of
walls, design of ceiling, etc.
d) Technology: The advanced technology will make your lives more comfortable. So, while
designing a good smart home, you should focus on security system, monitoring and
detec ng incidents, energy management, etc.
Now, you can think of some more provisions like a theatre room, rain dancing room, etc.
that you will like to have in your home. Here, we can provide some op ons of AI-powered
smart technologies that you can implement in your smart home:

Digital
Assistants

MY Fancy
Sensors SMART Ligh ng
HOME System

Cameras

Now, create a presenta on of the floor plan of your ‘Dream Smart Home’ using a
presenta on tool and present it in front of the class.

Artificial Intelligence - 9
18
 Programming With and Without Artificial Intelligence
As you know, a computer is a machine that accepts input
from the user, processes it and displays output. This input is
in the form of a set of instruc ons to do a par cular task,
known as computer programming. A er the existence of AI-
powered machines, we can differen ate programming in two
ways, i.e., Programming with AI and Programming without
AI. The detailed descrip ons of these two types of
programming are as follows:

A computer program with AI can answer the generic ques ons it is meant to solve. These
programs can integrate new modifica ons by pu ng highly independent pieces of informa on
together. Hence, you can modify even a minute piece of informa on about the program without
affec ng its structure.
A computer program without AI can only answer the specific ques ons it is meant to solve.
Modifica on in the program leads to change in the programming structure.

Fields Related to Artificial Intelligence

Nowadays, AI has become an important part of our lives. The field of AI has certain core
technologies that make it possible for machines to mimic the human brain and learning. The
core components of an AI-based system are as follows:
i) Machine Learning: Machine learning is a form of AI
that enables systems to learn automa cally from
experience or past informa on. For example, auto-
complete sugges on feature in search engines that
gives be er result with more experience and data.

ii) Deep Learning: Deep learning or deep neural learning is a

subset of machine learning. This system has the ability to
learn from unstructured or unlabeled data. For example:
Emoji Scavenger Hunt is a web browser-based game on
deep learning technique.

iii) Robo cs: Robo cs deals with the designing, construc on,
programming and applica on of robots. Robots are compu ng
machines which include both hardware and so ware. The Ar ficial
Intelligence technique is applied on the so ware aspect part of the
system to develop AI programs. The main aspira on in robo cs is
to use AI to develop AI programs and so ware for the computer to
perform ac ons and behave like humans. For example, a
collabora ve robot or a cobot can be easily programmed to pick up
objects and place them somewhere else.

PM Publishers Pvt. Ltd.

19
Applications of Artificial Intelligence
As we know, Ar ficial Intelligence is a cu ng edge technology that offers a series of important
benefits in many sectors, since it is one of the most versa le technologies that exist today. It
has become an essen al part of our daily life. Google, a voice assistant, is an example of
Ar ficial Intelligence that we use daily.
The experts of Ar ficial Intelligence are working on automa c learning, knowledge engineering,
ar ficial vision, natural language processing and data mining, to get really interes ng and useful
solu ons. Let us understand the applica ons of AI by following these points:

i) AI in Healthcare Industry: In the healthcare industry,

the main goal of implemen ng Ar ficial Intelligence
machines is to enhance produc vity in terms of
improvement in pa ent care and reducing work
pressure of the healthcare workers. To do this, many
healthcare industries have adopted Ar ficial Intelligence
machines to make be er and faster diagnosis.

ii) AI in Banking and Other Financial Ins tu ons: Ar ficial

Intelligence machines are used to collect personal data
and provide financial advice as per the requirement of
customers in banks and other financial ins tu ons. For
example, Eva, an acronym of Electronic Virtual Assistant is
India’s first and largest Ar ficial Intelligence banking
chatbot. This chatbot uses Natural Language Processing
(NLP) to understand the user query and fetch the relevant
informa on from a larger database within a second.

iii) AI in Business: Nowadays, most of the companies use a

biometric system for managing a endance of employees
effec vely and efficiently. Biometric system is an Ar ficial
Intelligence applica on that collects data on facial and
demographic characteris cs. Thus, intelligent machines
can adapt their responses by analyzing different iden es
and emo onal responses.

iv) AI in Manufacturing Industry: This is an area that has

been at the forefront of incorpora ng robots into the
workflow. AI-driven robots can work non-stop without
breaks, reduce opera ng expenditure and op mize
produc on of the manufacturing units.

Artificial Intelligence - 9
20