ASERC Journal of Socio-Economic Studies

Journal homepage: www.ajses.az Volume 6, Number 1, (2023) Pages 3-14

ARTIFICIAL INTELLIGENCE AND THE IMPACT

ON DEVELOPING COUNTRIES

Monikaa, Badri Narayanan Gopalakrishnana, Padma Iyengharb

ᵃ NITI Aayog, Government of India,
New Delhi, India.
b
University of Applied Sciences,
Osnabrueck, Germany

SERC

ABSTRACT

This paper examines the potential impact of Artificial Intelligence (AI) on

developing countries. The study focuses on the impact of AI on
employment, market structure, and international trade in these
countries. The analysis finds that AI has the potential to significantly
impact the labor market in developing countries, leading to job
A
displacement, wage stagnation, and the creation of new job
opportunities. It also finds that AI has the potential to alter market
structures and disrupt international trade, with significant implications
S
for the economies of developing countries. On, the one hand, AI and
robotics can erode the comparative advantage of the developing
countries in producing labor-intensive goods and on the other hand, AI
E
presents the opportunity to trade hitherto non-tradable services like tele-
medicine, digitalization etc. Finally, the study suggests that governments
in developing countries must take proactive measures to manage the
R
effects of AI, including investing in education and training programs,
promoting innovation, and establishing legal and regulatory
frameworks to protect workers and consumers.
C
Keywords: Artificial Intelligence, Automation, Job displacement, Job-
creation, Productivity, Super-star firms, Regulatory framework

3
 Monika, Badri Narayanan Gopalakrishnan, Padma Iyenghar

INTRODUCTION
The world is standing today at the cusp of another significant technological breakthrough
that will bring a paradigm shift in the way goods and services are produced. This
technological revolution is being heralded in policy and academic circles as the fourth
industrial révolution or as Industry 4.0. (Sangahvi,et al., 2019). With the exponential fall in
computing costs worldwide, it has become cheaper to employ more and more digital and
artificial intelligence technologies in the production of goods and services.
Industry 4.0 is based on rapid advances in information and communication technology,
thereby aiding intelligent networking of machines and processes for industry. Artificial
Intelligence (AI) can be considered as a backbone of the entire Industry 4.0 ecosystem as AI
algorithms permeate various facets such as optimizing the supply chain of manufacturing
operations, inventory control, staffing and energy consumption to name a few. Al refers to
the technology which makes machines capable of performing the tasks normally requiring
human intelligence, for example, speech recognition, natural language understanding and
processing and visual recognition. Naturally, humans are endowed with cognitive
capabilities, allowing them to do pattern recognition which machines could not do earlier
before the advent of Artificial Intelligence. To illustrate, let us assume that one person has
not seen a 'blue' coloured bowl in his life, but has seen a 'red' coloured bowl of the same type.
Now, if this person is directed to bring the 'blue' bowl, he would be able to perform the task
of bringing the blue bowl even though he has never seen it in his life. However, if the
description of the 'blue' bowl were not coded into the robot's data system, the robot would
have failed to execute the task. Thus, in the pre-Al age, humans had a comparative
advantage over machines in performing routine tasks requiring cognitive skills because of
their capability to do pattern recognition and to connect the dots from previous experiences
of similar types.
In recent times, extraordinary rapid advances are being made in Machine Learning (ML)
technology (Sangahvi et al., 2019). Improvements in this technology have enormous potential
to erode the comparative advantage of human beings in doing routine and repetitive tasks
based on simple pattern recognition. Machine Learning is the scientific study of algorithms
and statistical models that computer systems use to effectively perform a specific task
without using explicit instructions, relying on previous data patterns and inferences instead
(Rose, . et al., 2015).

1. ARTIFICIAL INTELLIGENCE AND ITS POTENTIAL IMPACT ON

PRODUCTIVITY AND EMPLOYMENT
According to Acemoglu and Restrepo (2018)), Industry 4.0 will result in the emergence of
technologies that can be categorized into two parts such as (1) Enabling technologies and (2)
Displacing technologies.
Enabling technologies augment the productivity of the person performing a job. For
example- The advent of the telescope did not replace the astronomers or scientists who were
earlier conducting space research; it merely augmented the productivity of these people in
performing their tasks.

4
 Artificial Intelligence and the Impact on Developing Countries

However, on the other hand, displacing technologies directly displace workers from their
current jobs; for example- milling machines and power looms directly took over the jobs of
the artisans. Thus, it becomes essential to distinguish whether the technology is substituting
or complementing the worker.
At the same time, it is also important to distinguish between jobs and tasks. A job is a
collection of different tasks which a worker does in his job. For example, a teacher's job
involves various tasks of delivering a lecture in a class on a specific topic, addressing
students' doubts about the delivered lecture, and customizing his/her approach to teaching
according to the different needs of the students.
The first two tasks can be automated leveraging Al and ML, but the third task (i.e.,
customizing the lecturer’s approach to teaching according to the different needs of the
students) will be more difficult to automate as it would be challenging to code it in terms of
the 'if and then' rules. However, even if we assume that with advances in future in this
technology, the third task also gets automated, even then, the teacher's job will not
disappear. Because apart from performing the above tasks, a teacher also provides various
sorts of counselling such as career and emotional counselling to students even when it is not
sought. This task involves more human attributes such as persuasion and empathy, which
may not be as effectively performed by a machine at this juncture as a human. Thus, it is
unlikely that a teacher's job will get displaced. Al can perform some tasks of teaching and
thus would end up increasing the productivity of teachers by allowing them to effectively
use their time and efforts on the more non-routine tasks requiring more human aspects.
In the direction of displacing technologies, let us consider some examples. For instance,
there exist some jobs which can be entirely automated with the onset of Al-enabled
production regime. For example, the job of IT professionals doing repetitive, rule-based,
routine coding can be completely taken over by Al-enabled technology (Ward et al., 2017).
And in this case, the technology has an upper hand and could render the workers
unemployed. Let us consider a more elaborate example. Even before the advent of AI, the
software development processes are undergoing a paradigm shift towards fourth generation
modeling languages, also referred to as the Model-Driven Software Engineering (MDSE).
With this emerging paradigm, the source code is automatically generated from higher level
models such as Unified Modeling language (UML) diagrams modelled by software architects
and engineers. Its practical applicability is gaining wide attention in the last two decades in
the industry, including the industry 4.0 and IoT applications as described and demonstrated
in Iyenghar & Pulvermueller (2018),. Now, with the advent of AI and a combination of AI
and MDSE, for instance AI-guided MDSE as described in Iyenghar & Pulvermueller (2022),
could further reduce, if not completely eliminate, the need for the IT professionals in the
future.
Now, let us consider the potential impact on productivity and employment. Besides the
displacement effect, either for some tasks or for the complete job, Al and robotics also entails
a productivity effect. A telling example of this process comes from the impact of the
introduction of automated teller machines (ATMs) on the employment of bank tellers. Bessen
(2016) documents that concurrent with the rapid spread of ATMs, a clear example of
automating technology that enabled these new machines to perform tasks that were
previously performed more expensively by labor, there was an expansion in the employment

5
 Monika, Badri Narayanan Gopalakrishnan, Padma Iyenghar

of bank tellers. Bessen (2016) suggests that ATMs reduced costs of banking and encouraged
banks to open more branches, raising the demand for bank tellers who then specialized in a
range of tasks that ATMs did not automate, like customer service, etc.
Also, suppose technological improvements result in technology deepening or increasing
capital productivity in tasks that have already been automated. In that case, it will generate
only productivity effects. These productivity effects then raise labor demand. (Acemoglu and
Restrepo, 2018)).
The substitution of expensive labour with cheaper machines or robots leads to significant
cost savings. The benefits of these cost savings can then either be passed on to the consumers
in the form of reduced prices or can lead to higher profit margins for the entrepreneurs. Let
us explore the potential effects that would arise via the first channel of reduced prices. If the
product in which technology has resulted in cost savings is characterized by higher price
elasticity, the overall effective demand for this product will rise. Also, if the fall in the cost of
production, and thus in the prices, is significantly large in the home country, then
international comparative advantage in producing the good may also arise, which will
increase the overall demand for the good for the producers in the domestic country.
Although employment elasticity in the production of this product falls, the total size of
employment may in fact rise due to expansion of production.
At the same time, a reduction in the price of the product also creates an income effect due to
which the demand for other goods or services may also rise and hence can increase the
employment opportunities in those sectors as well. Nevertheless, it is not necessary that the
cost savings are always transferred to the consumers in the form of reduced prices. They
might be retained by the entrepreneurs as well. The precise extent of sharing of cost savings
between producers and consumers depends on many factors, like market structure, market
power, contestability in the market, etc.
However, even if we assume that the entrepreneur retains the entire share of cost savings in
the form of higher profits, then also these increased profits will create incentives for further
investment in the sector. This will increase the demand for labour in cases where the tasks of
production of these goods cannot be automated. Nonetheless, the total effect on the induced
increase in the overall output can be expected to be lower via this channel. Because the
decision to invest by an entrepreneur also depends on the overall effective demand for the
good being produced. And in the scenario where the employees are getting laid off, a sense
of pessimism prevails amongst the households and the consumers, and this may reduce the
effective demand for all kinds of goods and services being produced in the economy. And
this may disincentivize entrepreneurs to invest further even when there is no lack of
available funds.
Besides, if Al personnel are in short supply, which is often the case in developing countries,
their scarce pool of talent would have to be paid wages much more than their marginal
productivity to make them stay longer with the firm. As a result, the firm may end up in a
situation where it is saving costs on the one hand by substituting labour by Al sophisticated
machines, and on the other hand, its costs get inflated due to the requirement of paying
wages to the few Al personnel. (Acemoglu and Restrepo, 2018). In this situation, the firm
may end up getting much less overall cost advantage. Furthermore, the above situation,
coupled with the depressed effective demand, might even cause recessionary pressures.

6
 Artificial Intelligence and the Impact on Developing Countries

Moreover, too much emphasis on automation may also crowd out the time available to the
common and the scarce pool of Al experts and scientists for other critical technological
breakthroughs. And this may also adversely impact the economy's overall productivity and
innovative activity.
Thus, any economy can only benefit by using Al, Robotics, and automation if and only if the
productivity effects are substantial. Thus, the real danger for the labour force does not come
from brilliant machines but from so-so machines (Brynjolfsson et al., 2020). These so-so
machines will be smart enough to displace the labour performing the routine tasks but will
not create that much productivity effect and might even lead to an absolute fall in
employment.
However, for the time being, let us assume that the productivity effects are significantly high
so that total employment ends up getting increased in the economy post-Al. Even in that
setting, the increase in output per worker will be more than the increase in wages (Acemoglu
and Restrepo, 2018)). Thus, even if we assume the productivity effects to be quite substantial,
even then, labour share in the total GDP is expected to fall.
Nonetheless, if we extend our static analysis to a dynamic setting, the effects of Industry 4.0
on the labour share may not look as gloomy. History has witnessed various technological
revolutions earlier as well. During agricultural mechanization, tractors also directly
displaced the farm labourers. The introduction of the steam engine and power looms also led
to the significant displacement effect (Mantoux, 1928). The question now arises: if all these
technologies have been able to displace workers successfully, how does one explain the
prevalence of so many existing jobs today? (Autor et.al, 2017).
The answer to the above question lies in the fact that periods of intensive automation have
often coincided with the emergence of new jobs, activities, industries, and tasks. New jobs
can emerge related to the current production process. For example- with the onset of Al, the
jobs for Al trainers may emerge. Not only in industry but also in universities, the need for
new courses and subjects pertaining to AI will increase substantially to address the market
needs. Also, there can be the emergence of the completely new jobs whose nature can't be
predicted at present. For example- A farmer in the 18th century would not have thought
about the emergence of products and services like computers, software products, electric
cars, banking services, etc. But these goods and services emerged and provided employment
to millions of individuals.
Similarly, the present Al breakthrough is also not expected to be entirely different from
history. However, it is important to note that this time, the penetration of new technologies is
expected to be much faster, and the real challenge lies in managing the pace of adjustment. It
is perhaps telling that wages started growing in the 19th-century British economy only after
mass schooling and when other investments in human capital expanded workforce's skills.
Similarly, the adjustment to the large supply of labor, which got freed from agriculture in
early 20th-century in America, was greatly aided by the 'high school movement,' which
increased the human capital of the new generation of American workers (Goldin &
Katz2009). The forces at work at present are likely to be more general than these examples.
New tasks tend to require new skills. But to the extent that the workforce does not possess
those skills, the adjustment process will be delayed. Even more ominously, if the educational
system is not geared to providing those skills (and if we are not even aware of the types of

7
 Monika, Badri Narayanan Gopalakrishnan, Padma Iyenghar

new skills that will be required so as to enable investments in them), the adjustment will be
greatly impeded.
Thus, without the intervention of the state, the mechanism of structural adjustment
highlighted above may be slow and can cause misery and lead to the painful costs of
adjustment for the millions of workers who get displaced from their jobs. This may lead to
social unrest, strengthening of interest groups, etc., which can adversely affect productivity
and further slacken the pace of adjustment.
Also, as opposed to the developed world, whose population is aging, the population
structure of many developing countries is displaying a youth bulge, and this phenomenon
will continue in the future. According to the statisticians, the only reason that the German
population did not decline in 2021 was the fact that more people are immigrating here than
emigrating abroad (Carter, 2022). For example, by 2030, India will have the youngest
workforce in the world (Kapila, 2016), further aggravating the country's employment
challenge. Thus, the success of Al crucially depends on how fast the infrastructural,
regulatory, and legal environment is made conducive for the Al-enabled production regime.

2. AI AND MARKET STRUCTURE

Now, we analyze the market structure in an Al-enabled production system. The advent of
different sophisticated technologies like cloud computing has significantly reduced the need
to incur fixed costs for maintaining data centers and servers. Cloud computing is the on-
demand availability of computer system resources, especially data storage and computing
power, without direct active management by the user (Sangahvi et al., 2019). Thus, this
characteristic of the market increases the market's contestability and helps put a restraint on
the incumbents' market power as any innovative entrant can come up with improved
product, production process, etc., without incurring too much fixed costs. Hence, this can
give rise to the startup culture.
However, at the same time, it should be noted that big global tech industries with deep
pockets also suck up Al talent from around the globe, leaving a shortage of human capital
skilled in Al elsewhere. These big companies can afford to pay very high wages to a few Al
personnel, which may lead to the concentration of Al talent in these superstar firms (Autor et
al., 2017). The significance of economies of scale and scope becomes even more pronounced
in an Al-based production system. It is worth having an Al team within a company if there
are a variety of applications for them to work out. Also, in contrast to the conventional
production system, big data is one of the key inputs in the Al-based production system, and
this big data is non-rival. It can be used simultaneously in multiple applications at the same
time without reducing its availability to any one of the other applications of a firm. Thus,
once the fixed costs for new digital services are covered, a growing market can be served at
approximately zero marginal cost.
Thus, in an era of digital technology, in which big data is a crucial input, big incumbent firms
have the ability to expand at an exponential pace. Besides, it is important to note that these
Al-based products and services are also characterized by significant network and lock-in
effects and thus further helping the incumbent firms to expand their market share (Autor et
al., 2017). Moreover, it is also feasible for these superstar firms to engage in predatory pricing
and bear losses to wipe out competition.

8
 Artificial Intelligence and the Impact on Developing Countries

Besides, the firms which have a large customer base, will be able to get more data on
consumers' behavioural patterns and while doing their econometric data analysis, as per the
law of large numbers, these firms would be able to come up with more useful and precise
estimates which will help them to further improve their customer experience and to get
further access to consumer data in return. As the competitive advantage of the incumbents is
reinforced, the power of the new entrants to drive the technological change may be
weakened.
Thus, it may be expected that although Al will reduce the entry costs for any potential
entrant, there will be more than ever significant and important first mover advantages in this
data driven Al-enabled production regime. The first mover firm will be able to capitalize on
its first mover advantage at a much faster pace in an Al-led production setting.
However, it must also be noted that the more personalized information about the
behavioural patterns of consumers will also enable the producers to charge more
personalized prices from consumers. The prevalence of third-degree price discrimination can
be assumed to be far more rampant in this new production system.
Since, big data will be one of the key inputs in the production process, sharing of this data
amongst firms also becomes a big issue. Private companies have strong incentives to
maintain data privately. However, if this non-rival data is shared amongst different players,
overall research output and the total quality and quantity of the goods and services
produced could be significantly higher. Thus, the market, on its own, will fail to reach a
socially optimal outcome because there exists a significant gap between private and social
incentives to share data outside of the formal institutional mechanism. In a game theoretical
framework, it will be socially optimal to share the non-rival data, but in the Nash
equilibrium each firm would end up keeping its data private unless any outside intervention
is made. Also, the private incentives for knowledge creation are naturally lower in
developing countries with a small market for technology-led products, poor capital market
backing innovation, and limited informational resources. Thus, the non-sharing of data and
less than optimal knowledge creation can also have adverse implications for the quality of
research and technical progress of developing countries and this can have perverse
ramifications on the global competitiveness of these countries in the production regime
where the first mover advantages loom large and are more than ever important in this new
setting. However, before addressing the question of global competitiveness, it is crucial to
understand the potential impact of Al on the world trade scenario.

3. AI AND INTERNATIONAL TRADE

With an exponential fall in the cost of computing and data storage, it becomes further
cheaper to communicate and share information globally. Furthermore, this time, this fall in
cost is coupled with rapid advancements in natural language processing and machine
translation. Thus, there arises a much wider scope to trade services globally, which was
hitherto considered non-tradable due to language and various other kinds of barriers. Al is
expected to accelerate the spread of global supply chains in which MNCs would integrate
their worldwide operations much more vigorously (Baldwin, 2019). Industry 4.0 will make it
feasible for a teacher in the US to sell his teaching services to the students in Kenya in their
own native language while sitting in his office in the US. Also, Al makes it feasible for

9
 Monika, Badri Narayanan Gopalakrishnan, Padma Iyenghar

Kenyan or Indian teacher to sell their teaching services abroad. For example, the integration
of advanced technologies such as artificial intelligence (AI), the Internet of Things (IoT), and
big data analytics can make online teaching experience very smooth by utilizing AI-powered
chatbots and intelligent tutoring systems to provide personalized assistance to students.
These systems can analyze data to identify areas of weakness or strength and develop
targeted learning plans. The use of IoT devices can also provide real-time feedback on
student progress, enabling educators to make adjustments and improve the learning
experience. Similarly, Natural language processing can enable the development of
conversational agents that can interact with students in real-time.
But what can be expected to happen to the trade competitiveness of developing countries in
producing the goods post-Al revolution? The onset of Robotics and Al-led production
regime will automate routine jobs, which can be easily coded in the 'if and then' rules. Thus,
under this setting of the production process, the share and importance of unskilled and semi-
skilled labourers in the production value chain are expected to fall substantially. And this
may erode the incentives for producers to move their activities from developed to
developing countries. Firms in developed countries may find it cheaper to automate certain
processes instead of offshoring. And it implies that the historical comparative advantage
enjoyed by developing countries due to their lower labour cost is at risk of getting eroded.
Thus, unless developing countries move up in the production value chain, there arises a
significant danger of workers getting displaced not only by automation but also by the
possibility of re-shoring. Countries like India, Vietnam, South Africa, and Morocco, which
have big outsourcing industries, mainly specialize in the back office administrative tasks in
banking, health, insurance, and accounting, which are at the highest risk of getting
automated.
Banga (2019) estimates the share of different countries in cross-border e-commerce and
highlights the losing trade competitiveness of developing countries and LDCs in digital
products. Thus, if developing countries fail to digitize their manufacturing, they may lose
their global trade competitiveness even in their traditional export sectors and domestic
markets.
However, the actual situation may not be as gloomy if developing countries leverage the
opportunity provided by Al to build their competitiveness in the hitherto non-tradable
services. There still exists a massive gap in the wages of service providers in developed and
developing countries, especially in those services that were not possible to trade in the pre-
Al era. Thus, developing countries must formulate the appropriate policy framework to train
and provide the required skills to their service providers to leverage this window of
opportunity. Otherwise, industry 4.0 may lead to further divergence in the income and
economic status of the developed and developing world.

4. ROLE OF STATE
The long-run strategy solution is to equip labourers with the required skills to enable them to
move up higher in the value chain so that Al-enabled technologies could complement their
skills instead of substituting them. And this requires the support of the state because markets
on their own may fail to create the optimal level of skilling and re-skilling (Zeira.,1998). And
this problem can be much more pronounced in the case of developing countries.

10
 Artificial Intelligence and the Impact on Developing Countries

However, this long-term solution to provide the appropriate skills to a large size of the
labour force, that too from a very narrow base, can take a very long time to manifest any
tangible outcomes. By contrast, technology and Al can penetrate the market and erode
labour cost advantages of developing countries at an extraordinarily fast pace. Thus, it might
become crucial to protect domestic industries from foreign competition. One such measure
that can be used is the data localisation policy. China has ingeniously exploited this policy
under the guise of stricter privacy norms (Vempati, 2016).
Data localization or data residency law requires that the data about a nation's citizens or
residents is collected, processed, and stored inside the country before being transferred
internationally. It will force foreign countries to set up their servers and data centers within
the host country's premises; thus, this can have significant positive technological spillover
effects on domestic firms. Furthermore, some crucial data sets about domestic consumers can
also be provided only to domestic firms. It would create an indirect subsidy to the domestic
Al industry. At the same time, foreign firms exporting their services to developing countries
must be encouraged to share their source code. Moreover, it is also imperative for
developing countries to preserve their policy space at the WTO negotiations.
However, it must also be noted that Al is a general-purpose technology, and government
must leverage this technology in governance. Al is the invention of the method of the
invention. Thus, the government can and must use this technology to plug various loopholes
in the monitoring, execution, and implementation of social welfare programs. Al can help the
policymakers constrained with limited fiscal resources in detecting tax frauds, preventing
subsidy leakages, and better target the beneficiaries in various social welfare schemes. Al can
be an asset of incalculable value in the predictive maintenance of public infrastructure. For
example, proactive physical infrastructure monitoring can be done using Al by combining
data points like usage metrics and image recognition to predict when and what maintenance
is required. It can prevent many accidents like train derailments, collapsing of bridges, dams,
etc. Al can also play a significant role in the agricultural sector as well. Precision farming has
been hailed as the next major agricultural milestone to improve farming productivity.
Further, in many developing countries, there is no proper way of monitoring the activities of
government officials and bureaucrats, which leads to massive corruption. Al enables the
government to check if things are happening on the ground (Sharma et al., 2020). Let us
assume that the bureaucrats in India are tasked with building toilets in public schools. In
India, to ensure that the toilets are actually built, the government has created a mobile app
where a government functionary will have to go to the toilet site, click a photo of the toilet,
and upload it to the central server. However, if there are one crore toilets to be built, this
would result in 1 crore photographs. But it would be impossible for the monitoring authority
to check whether each of these photos shows a complete toilet or a half-built toilet. If the
toilet in use, or is stashed with hay? Or if the 100 images have been uploaded by a
government functionary sitting in her office? So, the government usually relies on random
checks to create deterrence effects. But Al can create a way to process each of these one crore
photos and generate an alert whenever the photograph is not that of a wholly built toilet and
ensure that the multiple images don't get uploaded by a government functionary sitting in
her office. Similarly, the application of Al in education and health will not only lead to the
better and more efficient utilisation of public resources but can also help in bridging the

11
 Monika, Badri Narayanan Gopalakrishnan, Padma Iyenghar

rural-urban divide. Thus, if Al is leveraged in governance, it can help reduce the inequalities
in opportunities in any country.
Now let us assume that the government does not make any effort to leverage technology in
governance, or its pace of embracing Al is very slow compared to the rapid exponential rate
of penetration of Al in market-oriented sectors. Then, what would be the effects of Al on the
distribution of income and wealth under the above scenario?
In an Al-led market regime, the significance and relevance of old skills will become obsolete
much faster, and new skills will replace them. Al will incessantly replace workers from their
current jobs, and workers will have to update their skills to perform slightly modified tasks
continuously. If that entails requiring new skills that are costly to learn, automation can
increase inequality within occupations.
Economically disadvantaged individuals, for whom the burden of direct cost and the
opportunity cost of acquiring new skills is higher than their wealthier counterparts, will get
less training and may even move down in the work hierarchy. And the more prosperous
individuals will end up moving up the value supply chain in a scenario where no state
support is provided.
Besides, as routine tasks get automated, the significance and relevance of skills like creative
thinking, ability to innovate, having strong communication and leadership skills etc will
increase more than proportionately in the production value chain (ILO, 2018). However,
these skills require strong social, emphatic, and interpersonal competencies. These
capabilities can include innate ability, creativity, the ability to think out of the box, ability to
continuously persevere unless one gets the solution to the problem he/she is seeking. As
opposed to rule-based routine jobs, which are often based upon the knowledge of some rules
and tools, success under the Al-led production regime will largely depend on the inherent
personal capabilities of any individual.
It is important to note that mastery in future jobs will indeed require learning some new
rules and tools. But once these rules and tools are learnt, productivity will depend much
more on the innate ability of the individuals in an Al-enabled production regime. Hence, the
maximum productivity effect on the nation will occur when the individuals with the highest
innate abilities and creativities contribute to the national output.
However, the market equilibrium outcome can result in only richer individuals getting the
required skills and moving up in the production value chain. In the absence of government
support to learn and acquire skills for all, only the privileged individuals born into rich
families will have the resources to acquire the skill. No matter how talented, creative, and
innately intelligent individuals may be, without the opportunity to acquire the required skill,
they may not be able to contribute to the output according to their full potential. Hence, a
lack of opportunities to acquire the required skills will prevent many bright individuals from
deprived backgrounds to make the best use of their innate capabilities and talents. This will
ultimately get manifested in the aggregate level of productivity and competitiveness in a
country because, in the post-AI production regime, the comparative advantage will largely
be determined by creativity, interpersonal skills, ability to make judgements, etc. Thus,
inequalities in opportunities for training and education will not only have adverse impacts
on income distribution but can also adversely impact the overall productivity and, therefore,
the global competitiveness of any economy in an Al-enabled production regime.

12
 Artificial Intelligence and the Impact on Developing Countries

CONCLUSION
In conclusion, this research paper highlights the potential impact of Artificial Intelligence
(AI) on developing countries, specifically focusing on employment, market structure, and
international trade. The findings indicate that AI has the potential to bring about significant
changes in the labor market of these countries. While it may lead to job displacement and
wage stagnation, it also presents opportunities for the creation of new jobs. Moreover, the
study recognizes the transformative effect of AI on market structures and international trade.
On one hand, the adoption of AI and robotics may undermine the comparative advantage of
developing countries in labor-intensive goods production. On the other hand, AI offers the
potential for developing countries to engage in previously non-tradable services such as
telemedicine and digitalization.
To effectively manage the effects of AI, it is crucial for governments in developing countries
to take proactive measures. This includes investing in education and training programs to
equip the workforce with the necessary skills for the AI-driven economy. Promoting
innovation and providing support for research and development can also enable developing
countries to harness the potential benefits of AI. Additionally, establishing robust legal and
regulatory frameworks is essential to safeguard the rights of workers and protect consumers
in this rapidly evolving landscape. In nutshell, by adopting proactive strategies and
embracing the transformative potential of AI, these countries can navigate the challenges and
opportunities presented by this technology, fostering inclusive growth and sustainable
development.

REFERENCES
1. Acemoglu, D., & Restrepo, P. (2018). Artificial intelligence, automation, and work. In The economics of
artificial intelligence: An agenda (pp. 197-236). University of Chicago Press.
2. Acemoglu, D., & Restrepo, P. (2018). The race between man and machine: Implications of technology for
growth, factor shares, and employment. American economic review, 108(6), 1488-1542.
3. Autor, D., Dorn, D., Katz, L. F., Patterson, C., & Van Reenen, J. (2020). The fall of the labor share and the rise
of superstar firms. The Quarterly Journal of Economics, 135(2), 645-709.
4. Baldwin, R. (2019). The globotics upheaval: Globalization, robotics, and the future of work. Oxford University
Press.
5. Banga, R. (2017, June). Rising Product digitalisation and losing trade competitiveness. In United Nations
Conference on Trade and Development, https://unctad.org/system/files/official-document/gdsecidc2017d3_en.pdf
(Accessed 22.06.2023).
6. Banga, R. (2019). Is India Digitally Prepared for International Trade?. Economic and Political Weekly, 54(5), 2.
7. Belleflamme, P., & Peitz, M. (2018). Platforms and network effects. Handbook of Game Theory and Industrial
Organization, Volume II, 286-317.
8. Bessen, J. (2016) Learning by Doing: The Real Connection Between Innovation, Wages, and Wealth. New Haven.Yale
University Press.
9. Brynjolfsson, E., Benzell, S., & Rock, D. (2020). Understanding and addressing the modern productivity
paradox. Research Brief, MIT Work of the Future.
10. David, H. (2015). Why are there still so many jobs? The history and future of workplace automation. Journal
of Economic Perspectives, 29(3), 3-30.
11. Ernst, E., Merola, R., & Saman, D. (2018). The economics of artificial intelligence: Implications for the future of
work. ILO Future of Work Research Paper Series No, 5.
12. Goldin, C. D., & Katz, L. F. (2009). The race between education and technology. Harvard University Press.

13
 Monika, Badri Narayanan Gopalakrishnan, Padma Iyenghar

13. Iyenghar, P., & Pulvermueller, E. (2018). A model-driven workflow for energy-aware scheduling analysis of
IoT-enabled use cases. IEEE Internet of Things Journal, 5(6), 4914-4925.
14. Iyenghar, P., Otte, F., & Pulvermüller, E. (2022). AI-guided Model-Driven Embedded Software Engineering.
In MODELSWARD (pp. 395-404).
15. Kapila, U. (Ed.). (2008). Indian Economy since independence. 28th Edition, , Academic Foundation.
16. Mantoux, P. (2013). The industrial revolution in the eighteenth century: An outline of the beginnings of the modern
factory system in England. Routledge.
17. Manyika, J., Lund, S., Chui, M., Bughin, J., Woetzel, J., Batra, P., & Sanghvi, S. (2017). Jobs lost, jobs gained:
Workforce transitions in a time of automation. McKinsey Global Institute.
18. Sanghavi, D., Parikh, S., & Raj, S. A. (2019). Industry 4.0: tools and implementation. Management and
Production Engineering Review, 10(3), 3-13.
19. Sharma, G. D., Yadav, A., & Chopra, R. (2020). Artificial intelligence and effective governance: A review,
critique and research agenda. Sustainable Futures, 2, 100004.
20. Sirkin, H. L., Zinser, M., & Rose, J. (2015). The robotics revolution: the next great leap in manufacturing. BCG
Perspectives.
21. Vempati, S. S. (2016). India and the artificial intelligence revolution (Vol. 1). Washington, DC: Carnegie
Endowment for International Peace.
22. Ward, A. F., Duke, K., Gneezy, A., & Bos, M. W. (2017). Brain drain: The mere presence of one's own
smartphone reduces available cognitive capacity. Journal of the Association for Consumer Research, 2(2), 140-
154.
23. Zeira, J. (1998). Workers, machines, and economic growth. The Quarterly Journal of Economics, 113(4), 1091-
1117.

14