Artificial intelligence (AI) has become a critical tool in daily business operations worldwide, with applications spanning various fields. In India, as in many other countries, AI is increasingly integrated into sectors such as legal services, software development, and government departments. Despite the rapid adoption of AI technologies—from startups in Bengaluru to established corporations—economic indicators tell a different story. Productivity growth remains modest, GDP has not seen a significant uptick, and wages have not increased, suggesting that the anticipated “AI dividend” is lagging behind expectations.
The disparity between high AI adoption rates and modest productivity growth presents a puzzling scenario. In the United States, for example, labour productivity growth has stabilized at around 1–1.5 percent annually over the past decade, with no substantial acceleration following the emergence of powerful generative AI tools in late 2022. A working paper from the National Bureau of Economic Research highlights that nearly 28 percent of U.S. workers began using generative AI at work within 18 months of its introduction—an unprecedented rate of technological diffusion.
Intriguingly, while the individual effects of AI on productivity can be significant, the overall output of firms has not mirrored these gains. Research from MIT indicates that customer-support workers utilizing generative AI tools experienced an average productivity boost of about 14 percent, with newer employees seeing increases of 30–35 percent. However, companies chose to reinvest these efficiency gains into reducing training times and standardizing responses rather than expanding their output. As a result, overall firm output remained largely unchanged, despite clear improvements in productivity at the task level.
A similar trend was observed in a 2023 study by Harvard and the Boston Consulting Group involving management consultants. With access to advanced AI tools, consultants completed tasks 25–40 percent faster and delivered higher-quality work, yet their total billable output did not see a proportional increase. Firms absorbed productivity gains as time savings and reduced errors, leading to changes in internal workflows rather than growth in measurable output.
In India, prominent IT services firms have reported similar patterns, where AI tools have led to reductions in task completion times by roughly 20–30 percent in select service lines while overall revenue growth has remained subdued. This suggests that AI is primarily functioning as a mechanism to compress time rather than to enhance labour output, leading to faster decision-making and improved quality but not necessarily an increase in productivity as traditionally measured.
The limitations of current productivity statistics become evident when considering the nature of AI’s impact. These metrics were designed to evaluate output per hour worked but struggle to capture the nuances of how AI resolves uncertainty more rapidly. Historical precedents further illustrate this point. During the early era of electrification in the late 19th and early 20th centuries, productivity growth initially stagnated as industries adapted to new technologies without rethinking their workflows. Only after production processes were reorganized around electrical systems did productivity see a significant boost.
Economist Robert Solow famously remarked in 1987 that computers were visible everywhere except in productivity statistics. A similar narrative appears to be unfolding with AI, suggesting that it is reconfiguring work dynamics faster than it is translating into increased measurable output. This has direct implications for India’s economic landscape, as the nation increasingly relies on digital public infrastructure and services exports to stimulate growth.
As AI is deployed across various sectors, including software development, logistics, and government services, its early impact may manifest as faster workflows, improved coordination, and reduced errors rather than a marked increase in output per worker. Policymakers must be cautious; reliance solely on headline productivity figures may obscure the profound structural changes underway. There is a risk that policymakers might prematurely deem AI economically insignificant, which could lead to underinvestment in essential areas such as skills development and infrastructure.
Furthermore, if AI-induced efficiencies are absorbed internally by companies, the benefits may not translate into heightened wages or job growth, presenting a distributional challenge that requires careful policy consideration. The solution lies not in dismissing productivity statistics but in acknowledging their limitations. AI operates more like cognitive infrastructure—akin to electricity or the internet—than as a traditional labour-saving machine.
Ultimately, the significance of the AI era may not be heralded by a sudden surge in GDP, but rather through subtler indicators: quicker decision-making, fewer errors, and enhanced organizational efficiency. The real danger lies not in the failure of AI to transform economies but in the potential misinterpretation of its effects due to inadequate measurement systems. As AI continues to evolve, its quiet yet impactful contributions to productivity may reshape industries in profound ways, even if traditional metrics fail to capture these developments.
See also
AI-Powered Shopping Surges as 49% of Gen Z Use ChatGPT for Style Discovery
AI-Powered Fraud Detection Tools for 2026: Top 6 Solutions to Combat Document Forgery
Enterprises Adopt Agentic AI: 40% Surge in Applications Expected by 2026
NVIDIA Enhances AI PC Tools with 3x Faster ComfyUI and New LTX-2 Audio-Video Model at CES 2026
CryptoMarketForecast Launches AI Tool for Crypto Insights in Weather-Style Format
