6,000 Executives Exposed AI's Productivity Mirage — Here's What It Means for Your Job Security

If you have been worried that artificial intelligence is about to wipe out your job, a sweeping new study from the National Bureau of Economic Research (NBER) offers some reassurance. A survey of nearly 6,000 CEOs, CFOs, and senior executives across the United States, United Kingdom, Germany, and Australia found that close to 90% of firms reported no measurable AI impact on employment or productivity over the past three years. Despite billions pouring into AI adoption, most companies are using the technology for just 1.5 hours per week on average. Here at Metaintro, we track the data behind the headlines so you can make informed career decisions — and this study tells a story that is far more nuanced than the doomsday predictions suggest.

What Did the NBER Survey Actually Find?

The NBER working paper, authored by researchers including Ivan Yotzov and Jose Maria Barrero, surveyed business leaders at thousands of firms to measure AI's real-world footprint on jobs, output, and workplace productivity. The results paint a picture that challenges the breathless narrative around AI-driven disruption.

Approximately two-thirds of executives reported that their organizations are using AI in some capacity. Yet the average employee spends only about 1.5 hours per week using AI tools — a fraction of the typical 40-hour work week. A full 25% of respondents reported zero AI usage in their workplaces. When asked about AI's actual impact over the past three years, nearly 90% of firms stated it had produced no measurable change in either employment levels or productivity metrics.

This gap between AI investment and AI results has drawn comparisons to what economists call the "productivity paradox." As MIT economist Daron Acemoglu noted, the findings are "disappointing relative to the promises that people in the industry and in tech journalism are making." Apollo Chief Economist Torsten Slok put it more bluntly: "AI is everywhere except in the incoming macroeconomic data. Today, you don't see AI in employment data, productivity data, or inflation data."

Why Aren't CFOs Seeing the AI Payoff Yet?

The disconnect between AI hype and real-world results has multiple explanations, and they all matter for anyone navigating the job market in 2026. First, adoption is still shallow. While headlines focus on cutting-edge AI deployments at companies like Google, Microsoft, and Amazon, most businesses — particularly small and mid-sized employers — are still in the early experimentation phase. They have purchased licenses for AI tools but have not yet integrated them deeply into workflows.

Second, the NBER findings echo a well-documented pattern in technology adoption. Economist Robert Solow famously observed in 1987 that "you can see the computer age everywhere but in the productivity statistics." As Fortune reported, today's AI rollout is following a strikingly similar trajectory. New technologies typically take years — sometimes a full decade — before their productivity benefits show up in macroeconomic data, because companies need time to reorganize workflows, retrain workers, and build the supporting infrastructure.

Third, and critically for workers, the survey revealed that lowering costs — including labor costs — ranks among the least important motives for AI investment. Companies are not primarily buying AI to replace people. They are buying it to improve quality, speed up decision-making, and enhance the output of their existing workforce. This distinction matters enormously for anyone worried about mass layoffs driven by automation.

What Workforce Changes Are CFOs Actually Predicting?

While the headline story is about limited near-term disruption, the details of the survey reveal a more targeted reshaping of the workforce that job seekers should pay close attention to. CFOs are not predicting widespread job losses, but they are forecasting a meaningful shift in the types of roles companies need.

The most significant finding on the labor front: finance chiefs expect routine clerical roles to decline by 2 percentage points by 2028. These are positions built around repetitive, rules-based tasks — data entry, basic scheduling, standard reporting, and administrative processing. As AI tools become more embedded in daily operations, these tasks are among the first to be partially or fully automated.

On the other side of the equation, the share of technical roles — including engineers, data scientists, and analysts — is expected to grow by 0.62% in 2026 and by 1.35% by 2028. This may sound modest in percentage terms, but across the global economy, it represents tens of thousands of new positions. Companies that are investing in AI need people who can build, manage, maintain, and interpret these systems. The demand is not just for AI engineers; it extends to data analysts, project managers with technical fluency, AI trainers, and prompt engineers.

The overall employment forecast from executives is a net reduction of just 0.7% over the next three years. Individual employees surveyed were even more optimistic, anticipating 0.5% employment growth. Either way, these numbers are a far cry from the apocalyptic predictions of millions of jobs vanishing overnight.

How Does This Compare to What Tech Leaders Are Saying?

There is a fascinating tension between what this survey reveals and what Silicon Valley leaders continue to promise. Tech executives and AI developers have been vocal about AI's transformative potential, with some predicting that AI agents will replace entire categories of white-collar work within two to three years. The NBER data tells a very different story from the ground level.

As IT Pro reported, some tech industry leaders remain convinced AI will fundamentally reshape white-collar work within two years, even as the CFOs who manage budgets and headcount at thousands of companies say they have seen almost nothing yet. This disconnect matters because hiring decisions are made by the CFOs and CHROs, not by the AI evangelists. When the people controlling the purse strings say they are not planning mass layoffs, that is a more reliable signal than speculative predictions about AI's potential.

That said, the survey does show that executives expect AI's productivity impact to accelerate. The mean labor productivity growth attributable to AI was 1.8% in 2025, and firms anticipate this will strengthen in 2026 and beyond, with the largest effects concentrated in high-skill services and finance. The message is not that AI will never matter — it is that the timeline for meaningful workforce disruption is longer than many predicted.

What Does the Productivity Paradox Mean for Job Seekers?

For anyone actively searching for work or planning their next career move, this research carries several practical implications. The first and most important: do not panic. The data from 6,000 executives confirms that AI is not eliminating jobs at scale right now, and the companies doing the hiring do not expect it to do so in the near term. If you have been holding off on job applications because you assume your target role will be automated, this survey suggests you have more runway than the headlines imply.

However, "don't panic" does not mean "don't prepare." The shift away from routine clerical work and toward technical roles is real, even if it is gradual. Workers in administrative, data entry, and basic processing roles should be proactively building adjacent skills — particularly in data analysis, AI tool proficiency, and workflow automation. These are the capabilities that will keep you relevant as the 2% clerical decline plays out over the next two years.

As Metaintro CEO Lacey Kaelani told ZDNet, "A large number of our best product ideas have come from engineers doing the same repetitive data validation work over and over again, where they notice patterns that would lead to larger insight. However, once we eliminated that repetitive task and automated it, we definitely improved automation, but we lost the incidental learning that happens through seeing the data." This observation underscores a critical point: even when AI automates routine tasks, the human insight and pattern recognition that come from hands-on work remain irreplaceable — and employers know it.

Which Industries Will Feel the Impact First?

The NBER survey identified clear sectoral differences in where AI productivity gains are expected to land first. High-skill services — including consulting, professional services, and legal — along with the finance sector are projected to see the largest near-term effects. These industries deal heavily in information processing, document analysis, and pattern recognition, which are precisely the tasks where current AI models excel.

For job seekers targeting roles in financial services, this is a double-edged sword. The sector is likely to see both the highest AI-driven productivity improvements and the most significant reshuffling of job responsibilities. Entry-level analysts who currently spend their days building spreadsheets and pulling reports may find those tasks increasingly handled by AI copilots, while the demand grows for professionals who can interpret AI outputs, manage exception handling, and provide the strategic judgment that algorithms cannot replicate.

Manufacturing, healthcare, and retail — sectors that depend more heavily on physical labor and human interaction — are expected to see a slower and more limited AI impact on headcount. If you work in these industries, the near-term employment outlook remains relatively stable, though AI may change how you do your job even if it does not eliminate the job itself.

How Should You Position Yourself for What Comes Next?

The NBER data gives job seekers a clear strategic window. AI disruption is coming, but it is arriving gradually — which means you have time to prepare, but you should not waste it. Here are the concrete steps that align with what the survey reveals about where the labor market is heading.

Build AI fluency, not just AI awareness. The survey found that even at companies using AI, the average weekly usage is just 1.5 hours. Workers who can demonstrate genuine proficiency with AI tools — not just familiarity — will stand out. Take the time to learn prompt engineering, understand how AI copilots integrate into your specific field, and practice using these tools to produce measurable results. Include these skills on your resume with specific examples of how you have used AI to improve your work output.

Move up the value chain from routine to strategic work. With routine clerical roles on the decline, the smartest career move is to position yourself in roles that require judgment, creativity, and human interaction. If your current role is heavily administrative, look for opportunities to take on project coordination, client relationship management, or cross-functional collaboration — all of which are harder to automate and more valued by employers investing in AI.

Target the growing technical roles. The 1.35% growth in technical positions by 2028 represents real hiring demand. Data science, AI operations, machine learning engineering, and data analytics are all fields where employers are actively expanding headcount. You do not need a computer science degree to break into many of these roles — bootcamps, certifications, and demonstrated project work can open doors, especially when combined with domain expertise in industries like healthcare, finance, or logistics.

The Bigger Picture: AI as a Slow Burn, Not a Sudden Shock

Perhaps the most important takeaway from the NBER survey is the timeline. The executives closest to the budget decisions — the CFOs — are telling us that AI's impact on the workforce is real but slow-moving. The projected 1.4% productivity increase and 0.8% output increase over the next three years are positive signals for economic growth, but they are not the kind of numbers that translate into mass displacement.

This "slow burn" pattern is actually consistent with how every major technological shift has played out in modern economic history. The internet did not eliminate retail jobs overnight — it took over a decade for e-commerce to meaningfully reshape the sector. Personal computers did not replace office workers in the 1980s — they changed what office workers did. AI is following the same trajectory, and smart job seekers will use this transition period to evolve their skill sets rather than freezing in fear.

The global scope of the survey — covering the U.S., U.K., Germany, and Australia — also suggests this is not just an American phenomenon. Across developed economies, the pattern is the same: widespread AI adoption, shallow integration, minimal employment impact so far, and cautious optimism about future productivity gains. For workers in any of these markets, the strategic calculus is identical: prepare for change, but do not assume catastrophe.

People Also Asked

Q: Will AI replace my job in 2026?

A: According to the NBER survey of 6,000 executives, the vast majority of companies report no measurable AI impact on employment levels. CFOs forecast a net employment reduction of just 0.7% over the next three years, with the decline concentrated in routine clerical roles. Most jobs are not at immediate risk from AI, though the technology is gradually shifting what employers value in their workforce toward technical and analytical skills.

Q: What jobs are most at risk from AI automation?

A: The NBER data shows that routine clerical roles — including data entry, basic administrative processing, and standard reporting positions — face the most near-term pressure, with CFOs forecasting a 2 percentage point decline in these roles by 2028. Meanwhile, technical roles such as engineers, data scientists, and analysts are expected to grow. Workers in repetitive, rules-based positions should prioritize upskilling into areas that require judgment, creativity, or technical proficiency.

Q: How much are companies actually using AI right now?

A: While about two-thirds of firms report using AI in some capacity, actual usage is surprisingly low. The average employee spends only 1.5 hours per week on AI tools, and 25% of organizations report zero workplace AI usage. This gap between AI adoption (buying the tools) and AI integration (actually using them to transform work) is a major reason why productivity gains have not yet materialized at scale.

Future-proof your career by staying ahead of the data, not the hype. Metaintro delivers the research-backed labor market intelligence you need to make confident career moves — whether AI is reshaping your industry or not. Sign up today to get the latest workforce insights, job market analysis, and career strategies delivered straight to your inbox.