Artificial intelligence could lift aggregate labour productivity by as much as 3.8 per cent over the long term, according to a new International Monetary Fund (IMF) working paper that analysed patent and employment data across advanced economies. The study, published on Friday, 26 September 2025, titled Artificial Intelligence and Aggregate Labor Productivity: Evidence from Patent Data, offers some of the most detailed quantitative evidence yet on AI's economic impact.

What the IMF Research Found

The researchers used a production-function approach to assess whether the accelerating pace of AI patent applications translated into measurable gains in output per worker. Their findings confirm that the link is real: AI patent activity between 2000 and 2017 already raised output per worker by between 0.8 per cent and 1.2 per cent. The number of AI-related patents more than tripled over that period, with OECD countries accounting for roughly 89 per cent of all AI patents issued globally.

The estimated 3.8 per cent long-term productivity uplift is a projection contingent on continued development and broader cross-sector adoption of AI technologies. The paper stresses that such gains are not automatic — they depend on how quickly workers and businesses integrate AI into their workflows.

Who Benefits Most — and Why

The study found that economies with a higher share of employees in professional and managerial roles recorded larger productivity gains from AI. This suggests that AI delivers stronger economic returns when it complements skilled work rather than simply replacing existing tasks. Countries whose labour markets allow greater occupational mobility — enabling workers to shift toward roles where AI can be used most effectively — may capture a disproportionately larger share of those gains.

Notably, the benefits are unlikely to be distributed evenly across countries or employment categories, the paper cautions. Workers also need time to learn new systems, meaning productivity gains may emerge with a lag rather than immediately following adoption.

Methodology and Robustness

The researchers examined aggregate data from Organisation for Economic Co-operation and Development (OECD) member countries over the 2000–2017 period. Their conclusions held firm after controlling for technological spillovers — the mechanism by which AI advances developed in one country are adopted or adapted in others. The study's authors argue that patent creation alone is insufficient to maximise economic gains; countries also need a workforce capable of using the technology and labour markets that facilitate occupational transitions.

India's Stake in the AI Productivity Debate

India has placed growing policy emphasis on AI, digital public infrastructure, and technology-led economic growth. Its large information technology (IT) industry and substantial pool of skilled workers position it to benefit — but also mean that investment in training, upskilling, and workforce mobility will be critical as businesses expand AI deployment. The IMF findings add an evidence-based dimension to an increasingly urgent domestic policy conversation about how India captures productivity gains while managing potential disruption to workers and industries.

Globally, the debate over AI has shifted from questions of technological capability to its consequences for jobs, wages, and growth. Governments and multilateral institutions are now examining how investment in skills, research, and digital infrastructure can help societies capture the upside while protecting workers from displacement.