India not prepared for full force of artificial intelligence adoption

Only 11% of jobs in India can be automated against global average of 18%, says Goldman Sachs' assessment of 30 countries

Only 11-12 per cent of full time equivalent employment (FTE) in India can be automated, the lowest out of many countries, despite the growing popularity and adoption of artificial intelligence (AI) across the globe, according to a recent country-wise assessment by Goldman Sachs Global Investment Research.    

The global average of work which could get automated, including emerging and developed markets, is around 18 per cent, says the report. The study was based on 30 countries across the world, with eight developed markets taken into consideration. Goldman Sachs says that, based on its assessment, generative AI could expose over 300 million FTE across the globe to automation.           

The FTE is calculated by employee scheduled hours divided by employer hours for full time work. For example, if the employer’s full time work week  is 40 hours a week, employees who work the same are considered 1 FTE; if they work for 20 hours, they are calculated as 0.5 FTE.

 The impact of AI automation (including generative AI like ChatGPT) in India is the lowest, followed by Kenya, Vietnam, Nigeria, mainland China and Thailand where the impact on work which could be automated ranges from 13-16 per cent.

In contrast, the highest impact of work which could get automated by AI is in Hong Kong, followed by Israel, Japan, Sweden, the US and the UK where the percentage of jobs which could be automated ranges from around 25-30 per cent.

On the plus side, AI adoption is expected to boost annual productivity growth by a global average of 1.4 percentage points over a ten year period. Goldman Sachs estimates that this means a $7 trillion increase in annual global GDP over the 10 years. However, in India, the annual productivity growth is expected to be only 0.7 per cent over a 10 year period because the use of AI will be one of the lowest in the world.    

The assessment is significant. With the sudden popularity of AI-based ChatGPT, questions are being asked as to how many jobs and functions will become redundant and this has prompted fears of job losses.

Goldman Sachs estimates that the impact of AI on jobs will be felt more in developed economies rather than in emerging markets.

Its research has closely looked at the US and Europe and has come to the conclusion that at least two thirds of current jobs are exposed to some degree of automation and that generative AI could substitute one fourth of the current work.

Goldman Sachs has also given some insights on the kinds of jobs which are  more likely to be automated than others, based on detailed analysis of the US and Europe. In the US, 46 per cent of office and administrative staff work, 44 per cent of legal work, and 37 per cent of architectural and engineering work can be automated.

But it’s a very different story for other kinds of work. In building and grounds cleaning, only 1 per cent can be automated. For installation, maintenance, and repair work, only 4 per cent of work can be automated. In construction, the figure is 6 per cent. In these areas, automation can only complement existing jobs, not replace them.

How many jobs can be automated due to artificial intelligence?

India, Kenya, Vietnam, Nigeria, Mainland China, Thailand, Ghana, Indonesia, and Egypt are below the global average of 18 per cent of work which can be exposed to automation due to AI

21 countries which include the UK, US, Brazil, Mexico, South Korea, Euro area, Sweden, Japan, South Korea, among others, are above the global average of 18 per cent but below 30 per cent     

Productivity boost due to AI automation is higher than the global average of 1.44 percentage points annually over 10 years in 12 countries. Only Argentina is in the list from emerging markets; the rest are all developed markets  
 

Productivity boost due to AI is lower than the global average in 18 countries which includes only emerging markets