AI to reshape India’s workforce? 197 million workers face automation risk; 46 million may benefit: Study

A job exposed to AI contains tasks that artificial intelligence can perform, assist with or change. The article distinguishes between occupations with many routine tasks, which face possible substitution, and occupations where AI can improve productivity. Exposure therefore describes a job’s tasks, not an automatic prediction that the worker will be replaced.

For example, a routine-heavy role may include repeatable activities that AI can handle. By contrast, teachers, healthcare professionals, engineers and managers use expertise, judgement and responsibility. AI may support parts of their work while people remain important for decisions, communication and accountability.

The study estimates that 11.4% of India’s workforce faces substitution risk, while 7.6% could gain productivity. These figures show different possible outcomes. The report recommends reskilling, AI literacy and training in communication, critical judgement and problem-solving, rather than treating exposure as equal to job loss.

The study presents two different AI effects in India’s labour market. About 197 million workers are employed in occupations with a high share of routine tasks that could be automated. Separately, about 46 million workers are in jobs where AI could enhance productivity. The figures show that AI’s impact will not be uniform across workers.

The study estimates that 11.4% of India’s workforce faces possible job substitution. It also estimates that 7.6% could see productivity gains. Substitution means technology may perform some tasks that workers currently do. Productivity gains mean AI may help workers produce more or perform their work better, without necessarily replacing them.

These estimates come from India’s Periodic Labour Force Survey data. The report uses that national labour-market evidence to recommend early preparation. It calls for reskilling, stronger AI literacy and broader access to training that combines technical ability with judgement and problem-solving.

AI can complement a job when it strengthens a worker’s abilities instead of taking over the whole role. The article identifies healthcare professionals, teachers, scientists, engineers and managers as examples. Their occupations involve expertise, judgement and responsibility. These features make the work less like a fixed sequence of routine tasks.

A teacher might use AI to support parts of lesson preparation, while a healthcare professional could receive assistance with information or analysis. Engineers and managers may use technology to improve their work while continuing to make decisions and accept responsibility. The article does not claim that every task in these occupations is protected from automation.

The study found that Kerala, Haryana, Maharashtra and Tamil Nadu have the highest shares of workers in potentially complementary occupations. It recommends training that combines technical expertise with communication, critical judgement and problem-solving, helping workers use AI effectively while retaining essential human skills.

The study identifies four states with the largest shares of workers in occupations where AI is more likely to complement human labour. Kerala ranks first, followed by Haryana, Maharashtra and Tamil Nadu. In these occupations, AI could assist workers and improve productivity rather than necessarily replace them.

Kerala has 14.4% of its state workforce in such roles. Haryana has 11.5%, Maharashtra has 11.1% and Tamil Nadu has 10.7%. The occupations include healthcare professionals, teachers, scientists, engineers and managers. Their work depends on expertise, judgement and responsibility, which allows technology to support human effort.

These figures describe the share of each state workforce in potentially complementary occupations, not total AI adoption. The report says states should prepare before AI’s effects become visible in employment data. It recommends reskilling, AI literacy and training that combines technical skills with communication, judgement and problem-solving.

When a job contains many routine tasks, AI may be able to perform some of those tasks. That can reduce the amount of work available to people in the same form. The article describes this as a risk of job substitution, and it highlights the possibility of worker displacement in some occupations.

The study places nearly 197 million Indian workers in occupations with a high share of routine tasks that could be automated. This does not mean every one of these workers will lose a job. It means their work contains activities that may be especially exposed to automation. The actual effect can differ across occupations and workers.

The report recommends preparing before job losses appear in employment data. Its proposals include targeted reskilling, stronger AI literacy and training in technical expertise, communication, critical judgement and problem-solving. Such preparation could help workers move toward higher-productivity employment as job tasks change.

Uttar Pradesh, Bihar, Jharkhand and Madhya Pradesh face a different starting point from Kerala, Haryana, Maharashtra and Tamil Nadu. They have lower exposure to occupations where AI could improve productivity. They also have fewer workers in roles where technology can complement human skills.

That creates a dual challenge. Workers may have fewer opportunities to gain from AI, while the states may also have fewer occupations that combine technology with human expertise. The article therefore points to a need for pathways into higher-productivity employment, rather than focusing only on protecting workers from substitution.

The report recommends wider access to training and targeted reskilling. It also calls for expanding healthcare, education, elderly care and skilled trades, where India faces labour shortages and human interaction remains important. The stated priority is to prepare workers before AI’s employment effects become evident, not after losses materialise.

In basic economic terms, productivity measures output compared with the resources used to produce it, such as workers and their time. Higher productivity means producing more goods or services with the same number of workers, or producing the same amount with fewer resources. The article uses productivity gains to describe one possible benefit of AI.

AI could improve productivity by assisting workers with parts of their jobs. A teacher, engineer, healthcare professional or manager might use technology to support work that still requires expertise, judgement and responsibility. The worker remains involved, while AI helps increase the amount or quality of output. The source does not give a specific example of an AI tool.

The study estimates that 7.6% of India’s workforce could see productivity gains from AI, or about 46 million workers. It recommends technical training alongside communication, critical judgement and problem-solving so workers can benefit from this potential.