A study released by Cognizant on November 3 estimates that AI-enabled workforce productivity could contribute $28 billion to $84 billion in additional annual nominal GDP under realistic near-term adoption
scenarios. The potential is considerably larger if AI becomes widely accessible across the workforce.
However, the findings also reveal a disconnect between employees experiencing productivity improvements and companies seeing tangible business returns. While 97% of employees surveyed in India reported some productivity improvement from AI, 84% of business leaders said their organisations had not yet seen AI reduce costs. Another 91% reported no revenue impact.
The findings raise a key question: If AI is helping employees work more efficiently, why are companies still struggling to translate those gains into profits?
What The $84 Billion Opportunity Really Means
Cognizant’s research estimates that AI-enabled workforce productivity could add between $28 billion and $84 billion to India’s annual nominal GDP, depending on the extent of adoption and how effectively productivity improvements translate into economic output.
The study also identifies a theoretical upper limit of $830 billion annually. This assumes widespread AI adoption across the entire workforce and that productivity gains fully translate into economic output.
The $830-billion figure, however, is not a forecast of what India will earn from AI. It represents a theoretical scenario based on much broader adoption and the full conversion of productivity improvements into economic value.
The more realistic $28-billion to $84-billion range is also modelled rather than a guaranteed addition to India’s GDP.
The estimates are based on a survey of 700 enterprise employees and 75 C-suite and business executives across 10 major industries. The companies surveyed reported revenues ranging from $100 million to more than $10 billion.
The study suggests that India has the workforce readiness and infrastructure to capture a significant share of this opportunity. But the actual economic impact will depend on how quickly businesses deploy AI and whether they can turn employee-level efficiency into measurable output.
Employees Are Seeing AI Gains, But Access Remains Limited
The strongest signal from the report is the productivity improvement reported by employees. Nearly 97% of respondents said AI had improved their productivity, with the average reported improvement standing at 15%.
This suggests that AI is already helping workers perform at least some tasks more efficiently. However, the findings also expose a gap between employee preparedness and enterprise-wide adoption.
Around 74% of employees surveyed in India had completed AI training, compared with 56% globally. Yet business leaders estimated that only 36% of employees, on average, had access to AI tools.
In other words, employees are preparing for AI, but access to the technology remains restricted across organisations. This gap shows productivity improvements among a limited group of employees may not be enough to transform an entire business. Companies need to extend access beyond small teams and individual experiments if they want AI to influence operations at scale.
“Just because productivity at work has increased, does not mean there are any returns on the investment that an organisation has made in AI. An employee can complete his work faster, but unless the organisation redesigns processes or improves the way in which it delivers services, there will be no impact on the top line or bottom line. EY India’s research found that in enterprises where GenAI workloads are in production, only 8% of companies are able to fully measure and allocate costs associated with AI. Without measurements and performance indicators, AI is a waste of investment,” said Kumar Rajagopalan, Vice-President, Strategic Initiatives & Country Head India, Dexian.
Why Productivity Gains Are Not Translating Into Profits
The biggest challenge is converting faster work into financial value. According to the Cognizant study, 35% of leaders reported workforce productivity gains. However, 84% said their organisations had not yet experienced AI-driven cost reductions, while 91% reported no impact on revenue.
Another 37% of leaders said they had paused or stopped AI initiatives because of unclear returns on investment. These findings highlight the difference between making employees more productive and improving a company’s financial performance.
“For many firms, the introduction of AI is not a revolution in their business models; it’s an extra productivity capability to enable the business to run more efficiently. The savings from the benefits – that is, reduction in turnaround time or enhanced research – are not guaranteed to be offset by reductions in operating expenses or increased operating profits,” said Kanishk Agrawal, Founder & CEO, Path in AI Technologies.
AI may help employees produce more work without immediately generating additional revenue. The financial benefit depends on whether businesses can use those improvements to serve more customers, develop new offerings or improve the efficiency of their operations.
The report’s findings suggest that companies are still working out how to connect AI deployment with measurable business outcomes.
“An employee can complete his work faster, but unless the organization redesigns processes or improves the way in which it delivers services, there will be no impact on the top line or bottom line. EY India’s research found that in enterprises where GenAI workloads are in production, only 8% of companies are able to fully measure and allocate costs associated with AI, said Rajagopalan.
Healthcare Leads, While Manufacturing Faces A Challenge
AI productivity gains are not uniform across industries. Healthcare emerged as the leading sector in the survey, with 38% of employees reporting productivity improvements above 20%.
Communications followed, with 31% reporting gains above 20%. Technology and banking recorded 28% each, while retail and consumer packaged goods reported 18%.
Manufacturing had the lowest share of employees reporting any level of productivity improvement, at 82%. However, 27% of manufacturing employees expected productivity gains above 20%.
The variation suggests that AI’s impact depends partly on the nature of work and how easily the technology can be incorporated into existing processes.
“The time saved by GenAI can be put into productive use in healthcare, banking, and the technology sector, amongst others. The 8-10 hours a week saved by knowledge workers, across various industries, would translate into GDP growth in the country, provided there is wide access to AI, digital infrastructure to support it, a skilled workforce to work with it, and responsible AI usage,” said Rajagoplan.
The findings also indicate that AI adoption should not be measured through a single productivity figure across industries.
What India Needs To Unlock AI’s Economic Gains
India’s technology infrastructure could support wider AI adoption. Around 79% of senior leaders surveyed said their organisation’s technology infrastructure was accelerating AI rollout, while 68% viewed India’s national infrastructure as an accelerator. But infrastructure alone will not determine whether AI delivers economic value.
Businesses will need to expand employee access to AI tools, continue investing in training and deploy the technology in more complex, higher-value work.
“Enterprises should pinpoint their issues, set up measurable metrics for success, and examine the impact of AI investments on various business metrics, such as revenue growth, cost savings, customer experience, and operational efficiency,” advised Agrawal.
The next priority should be strengthening data infrastructure, governance and employee capabilities. “The key to AI success isn’t about how many we deploy, it’s about the value that we deliver and the outcomes that matter to the business,” he added.
Rajagopalan quoted a Deloitte research from 2025 that said, in most cases, only 6% of results obtained from typical use cases of AI bring returns within a year or less. Most often, the payback period for such investments is between two and four years. Thus, there is a need to “continuously” invest in employees, in quality data, and in responsible AI governance in order to find out what works best, to identify the gaps, and to scale up accordingly.
