Last Updated:
AI hype is soaring, but results remain limited. A landmark survey of 6000 executives highlights that most firms see no significant impact, while millions of jobs face uncertainty.
AI hype is everywhere, but impact is scarce; Over 90% of firms report no measurable job or productivity gains, even as future disruption looms. (Image-AI)
Billions have been poured into artificial intelligence, and there are almost no visible results. A sweeping global survey now challenges the AI revolution and focuses on the widening gap between AI promises and performance.
Yet, as Abhishek Ranjan, CTO of Blue Machines AI, points out, “A lot of spend is foundation, compute, data plumbing, integrations, security and compliance, which looks like cost before value shows up.”
The global race to embed AI into workplaces is becoming a popular topic for every sector. From boardrooms in New York to financial districts in London and manufacturing hubs in Germany, executives have championed AI as the next engine of growth.
Headlines have promised leaner teams, significant productivity gains and smarter automation, but beneath this optimism lies a more sobering story in the data.
A comprehensive study published earlier this month by the National Bureau of Economic Research (NBER) offers the clearest data on how AI is shaping firms across advanced economies.
Surveying nearly 6000 CEOs, CFOs and senior executives across the United States, Germany, the United Kingdom and Australia, the researchers sought to answer an urgent question: Is AI delivering measurable economic impact?
What The NBER Study Says
The NBER paper, described as the first representative international firm-level survey of AI adoption based on responses from senior executives, was conducted between November 2025 and January 2026.
Research teams from the Federal Reserve Bank of Atlanta, Bank of England, Deutsche Bundesbank and Macquarie University coordinated parallel surveys to ensure consistency across countries.
“On average, more than 90 per cent of business managers across the four countries estimate no impact of AI on their employment over the past three years. 89 per cent report no impact of AI on their labour productivity (measured as volume of sales per employee) over the last three years,” the study highlighted. Despite widespread adoption, measurable gains remain elusive.
Adoption Is High, But Usage Is Shallow
Around 70% of firms reported actively using AI technologies. Yet the average executive usage is limited to just 1.5 hours per week. Roughly 25 per cent report no use at all.
The Integration Gap Holding AI Back
Abhishek Ranjan offers a counterintuitive take. “1.5 hours per week is actually high compared to many SaaS tools that people use in short bursts and still get massive value from. AI is increasingly embedded inside the tools we already use, IDEs, docs, email, CRM, so time spent in a standalone chat box undercounts real usage.”
In his view, the real shift happens when AI interactions translate into “controlled workflow outcomes like compliant summaries, safe writebacks, automated QA, follow-ups and escalations inside the system of work.”
But Anil Agarwal, CEO of InCruiter, disagrees sharply.
“Honestly, no. 1.5 hours per week is tool exploration, not workflow transformation,” he says. “They’ve licensed AI. They haven’t integrated it.” Agarwal argues that AI only delivers when it is woven into the core workflow. In recruitment, for instance, that means AI operating inside the interview room, scheduling queue and feedback loop, not as a separate dashboard.
Ankit Aggarwal, founder of Unstop, sees the same pattern. “Usage of one to one and a half hours a week reflects experimentation, not transformation. Meaningful adoption happens when people do not consciously think about using AI. It simply becomes part of how they hire, analyse data, make decisions or learn new skills.”
The divergence in opinion reflects a broader uncertainty: Is AI underused or simply mismeasured?
The Productivity Puzzle
AI’s promise has always rested on productivity, automating repetitive tasks, reading and analysing massive datasets and generating reports instantly. Yet 89% of firms report no measurable productivity gains over the past three years.
“Despite $30-40 billion in enterprise investment into GenAI, this report uncovers a surprising result in that 95 per cent of organisations are getting zero return,” the MIT study noted.
Ranjan argues that the returns are visible, but in narrow domains first. “The first place clear returns are visible is coding and developer workflows because the environment is structured and measurable. Enterprises are next, once guardrails, auditability and permissions are in place.”
Agarwal believes the issue is measurement. “Most organisations are measuring the wrong thing. They’re looking at incremental efficiency gains in isolated tasks. The return shows up in time to hire dropping from 45 days to 6, in quality of hire improving, and in recruiter bandwidth freed up.”
Meanwhile, Ankit Aggarwal highlights organisational readiness as the real bottleneck. “AI itself is not the constraint. The barriers are data maturity, fragmented legacy systems and human capability. Until workflows are redesigned around AI, ROI will remain premature.”
No Job Losses Yet, But 1.75 Million Jobs At Risk
On employment, the NBER findings are equally nuanced. Over 90% of managers report no impact on jobs in the past three years. Yet executives forecast that over the next three years, AI will:
- Boost productivity by 1.4%
- Increase output by 0.8%
- Reduce employment by 0.7%
“Given over 250 million people are employed in employment over these four countries, firm executives therefore expect AI will lead to about 1.75 million fewer jobs by 2028 at existing firms,” the study stated.
Yet, employees surveyed separately predict a 0.5% increase in employment. Agarwal views the job-risk figure as both a warning and an opportunity. “That risk is real, but so is the opportunity. The organisations that will lead are those that stopped treating AI as a productivity experiment and started treating it as a core infrastructure decision.”
Ranjan frames it as evolution, not elimination. “When AI becomes ambient and default across devices and apps, turning intent into action instantly—that is when adoption moves from weekly usage to daily dependence.”
Ankit Aggarwal adds that the workforce impact depends on preparedness. “The long-term value of AI will depend less on how much is spent on technology and more on how deeply companies invest in people, skills and organisational change.”
Why The Impact Has Been Limited So Far
Across expert responses, several themes emerge:
- Integration Over Installation: AI tools are often deployed as add-ons rather than embedded systems.
- Governance And Guardrails: Ranjan stresses compliance barriers: “AI must safely read, write and act inside systems of record, CRM, ticketing, approvals, with PII controls, audit trails and clear accountability.”
- Leadership and Change Management: Agarwal calls the productivity gap “a leadership problem, not an AI capability problem.”
- Organisational Readiness: Ankit Aggarwal emphasises that fragmented systems and unstructured data make seamless integration difficult.
Hype Cycle Or Structural Shift?
Media attention around AI has surged nearly twentyfold since 2020. Yet the NBER study underscores the value of representative, cross-country data drawn directly from senior executives rather than anecdotal reports.
The findings suggest that AI is widely present, but not yet structurally embedded. Historically, transformative technologies, from electricity to the internet, required complementary infrastructure before productivity gains became visible. AI may be following a similar path.
Executives remain optimistic. Nearly 75 per cent of businesses anticipate adopting some form of AI within the next three years.
Should Workers Be Worried?
The projected 1.75 million jobs at risk trigger anxiety. But experts urge nuance. The forecasted 0.7 per cent employment reduction over three years is modest relative to past industrial shifts. Moreover, employees themselves expect net job creation.
The more likely outcome may not be mass unemployment, but job redesign. Routine tasks may shrink. Strategic and oversight roles may expand. Workers who learn to collaborate with AI systems could find themselves augmented rather than replaced.
Artificial intelligence commands headlines, capital, and executive enthusiasm. Yet, the latest evidence suggests that, for now, its workplace revolution remains aspirational.
Over 90% of firms report no productivity gains, no significant job losses have occurred, usage remains limited, but optimism persists. As Ranjan notes, much of today’s spending is foundational. As Agarwal argues, leadership intent determines outcomes. As Aggarwal stresses, preparedness and skills will shape the future.
For workers and firms alike, panic may be premature. But transformation, when it comes, will reward those who move beyond experimentation and into intentional redesign. AI’s true impact may not lie in today’s hype, but in tomorrow’s integration decisions.
February 19, 2026, 13:22 IST
Stay Ahead, Read Faster
Scan the QR code to download the News18 app and enjoy a seamless news experience anytime, anywhere.
