Is global economy ready for AI?

The IMF report says that the exposure of global employment to AI is almost 40 per cent.

Artificial Intelligence (AI) is no longer a topic for sci-fi novels and movies. It is now becoming part of common parlance with political and business leaders discussing the potential impact of AI on economies and businesses. The discussions at the recently concluded World Economic Forum in Davos centred on AI.

Amidst the heightened discussions, economists at the International Monetary Fund (IMF) released a report analysing the impact of AI on the future of work. The report analyses the impact of AI on country-wise and sector-wise employment.

The IMF research says that the exposure of global employment to AI is almost 40 per cent. This means nearly 40 per cent of global jobs are under pressure. Within the global economy, the exposure of advanced economies, emerging economies, and low-income economies is 60 per cent, 40 per cent, and 26 per cent respectively.

Despite the projected adverse impact of AI on unemployment, economies are encouraging the usage of AI to boost productivity and economic growth. Developed countries, in particular, have stagnated in terms of productivity, and are looking at new engines of growth. There are expectations that AI could be the next engine of growth. The IMF research notes that even though advanced economies are likely to be impacted more adversely than emerging economies, they are also more likely to seize AI’s advantages than emerging economies.

In most of the other technological revolutions, the impact is usually on the low- and middle-income jobs. The one striking aspect of AI is that it is likely to impact high-income jobs. AI’s advanced algorithms can analyse large-scale data, process images, recognise patterns, and take quick decisions. These tasks earlier were done by high-skilled people who were immune from the previous cycles of automation. Not this time around.

IMF economists have developed a framework to measure the impact of AI on the workforce. They use two criteria of exposure to AI and complementarity to AI for categorising occupations. Exposure to AI is defined as the degree of overlap between AI applications and required human abilities in each occupation. Complementary means AI could be used as an additional tool to help occupations.

Based on the two criteria, the IMF has categorised occupations in three categories. The first category of ‘high exposure, high complementarity’ includes professions which are threatened by AI but society is unlikely to allocate the tasks to AI. For instance, even though AI can do the tasks of judges, surgeons etc., society will not be comfortable — at least for the time being. Hence, these professions could use AI for better judgment delivery and precision surgeries.

The second category is ‘high exposure, low complementarity’, which comprises occupations which have high exposure but are likely to be replaced. For instance, telemarketers, call centres, etc.

The third category is ‘low exposure’ where the jobs have limited or no scope for AI application. This comprises people who do tasks with hands such as cleaning, carpentry, and so on.

The framework shows that advanced economies have the highest share of ‘high exposure’ occupations, followed by emerging economies and low-income economies. On the other hand, low-income economies have the highest share of the workforce that has ‘low exposure’ to AI.

The employment shares of select economies shows that the United Kingdom exhibits ‘high exposure and high complementarity’ as it has a significant portion of employment in professional and managerial occupations. The United States has a lower share of high exposure occupations, but it is still high. The other three countries: Brazil, South Africa, and India have a much lower share of high AI exposure professions. As per the report, in India, most workers are craftspeople, skilled agricultural workers, and low-skilled, or ‘elementary’ workers; most of these are in the low-exposure category.

In terms of gender-wise differences, in most countries, more women are employed in high exposure occupations. This provides both opportunities and challenges for women. The research also shows that the younger population and those with college education are more likely to transition to AI-enabled jobs.

To sum up, AI has opened a Pandora’s Box for the world economy. AI is expected be a sign of both high hopes and high misfortune. In most of the previous technological revolutions, the high income economies and high-end sectors gained, whereas low income and low end sectors lost. AI poses greater challenges to high-income economies and high-end sectors—however, they are also more likely to transition compared to their lower counterparts.

India with its agricultural and low-skilled workforce is not highly exposed to AI, but we will not be able to make exponential gains from the AI revolution. The challenge facing all economies is how to lower the risks from AI in terms of employment and make gains in terms of growth. These are still early days, but the current evidence suggests that humanity will have to carefully unleash AI as it changes every aspect of our lives.

(Amol Agrawal is an economist teaching at Ahmedabad University.)

Disclaimer: The views expressed above are the author's own. They do not necessarily reflect the views of DH.