‘I’m Sorry I’m Letting You Go Because We’ve Introduced AI In Our Business’

“What a great headline: ‘I’m sorry I’m letting you go because we’ve introduced AI in our business.’ It’s just much more kind and gentle than saying, ‘I’ve kind of employed you for the last three years but I don’t really need you.’”

Ken Griffin made that remark in an interview with The Wall Street Journal, discussing how companies explain layoffs at a time when artificial intelligence is widely discussed but unevenly deployed. He said companies increasingly cite AI when announcing layoffs, even when other factors are driving the cuts.

That is accurate. Companies across the U.S. have reduced headcount following aggressive hiring during the pandemic and a subsequent slowdown in demand. In public communications, many have pointed to AI as a factor behind those decisions. Labor and productivity data indicate that, in most cases, layoffs are better explained by post-pandemic normalization and cost control than by realized productivity gains from automation.

Why AI Gets the Blame for Layoffs

Griffin described recent layoffs as a correction following pandemic-era labor hoarding. “The employment market today is still reasonably robust,” he said, “but it’s not as tight as it was two years ago.” Companies, he added, are trimming roles that are not considered strategic.

Layoff data aligns with that explanation. Challenger, Gray & Christmas reported that U.S. employers announced 108,435 job cuts in January 2026, the highest January total since 2009. The firm cited restructuring, market conditions, contract losses, and store closures as the most common reasons. Only 7,624 layoffs that month (about 7 percent of the total) were explicitly attributed to AI.

For all of 2025, Challenger tracked 54,836 job cuts tied to AI, while total announced layoffs for the year exceeded 1.2 million. AI-linked cuts therefore accounted for a small share of overall reductions, even as AI appeared frequently in company explanations.

Business research has noted the same pattern. Harvard Business Review reported that companies increasingly reference AI’s expected future impact when explaining layoffs, rather than citing productivity improvements already in place.

Referring to AI allows companies to explain workforce reductions without describing roles as excess or the result of earlier overhiring. The explanation aligns with current expectations around technology investment, even when the underlying drivers are broader economic pressures.

What the Productivity Data Shows So Far

Griffin also questioned whether AI is delivering productivity gains at the scale implied by layoff announcements. “Objectively,” he said, “very few businesses are actually seeing productivity gains that come anywhere close to the headline of job losses that we have seen.”

Aggregate productivity data supports a cautious reading. U.S. nonfarm business productivity increased in 2024 and 2025, according to the Bureau of Labor Statistics, but gains were modest and uneven across industries. The data does not show a broad acceleration consistent with widespread, automation-driven efficiency gains.

Firm-level research points in the same direction. A 2025 Boston Consulting Group survey found that about 5 percent of companies reported significant, measurable value from AI investments, defined as sustained improvements in revenue, cost structure, or productivity. Most companies remained in early deployment stages or pilot programs

While employees often report time savings from AI tools, a substantial portion of those gains is offset by rework, including correcting errors, verifying outputs, and rewriting content. In a Workday survey, roughly 40 percent of reported time savings was lost to rework, reducing net productivity impact

Academic studies show similar outcomes. Research on large language models finds that AI can improve speed or output quality for certain tasks, particularly for less experienced workers. However, those gains are task-specific and do not consistently translate into higher productivity at the team or firm level without changes to workflows, training, and management practices

Taken together, the data helps explain the gap Griffin described. Layoffs have taken place while many firms are still experimenting with AI rather than deploying it at scale. Productivity improvements exist, but they remain limited and uneven. At present, they do not match the scale of workforce reductions that cite AI as a contributing factor.