The AI Layoff Story Has a Plot Hole

In October 2025, Mike Cannon-Brookes, the CEO of Atlassian, went on a podcast and told the world his company would employ more engineers in five years, not fewer. He was confident, specific, and on the record.

Few weeks ago, Atlassian announced it was cutting 1,600 people, over 900 of them from software research and development. The reason given: artificial intelligence.

The sequence felt familiar. In the way that something feels familiar when you’ve watched the same pattern play out before, just with different names and different numbers.

The past few months have produced a remarkable volume of AI-attributed layoffs. Block cut 4,000 people, roughly 40% of its entire workforce. Jack Dorsey wrote in a shareholder letter that intelligence tools had changed what it meant to build and run a company. Dozens of other companies have filed similar announcements, each one positioning artificial intelligence as the primary driver of structural change.

The numbers are real. The human cost is real. But ... may be should look at it deeper.

Forrester Research, in its 2026 workforce predictions, published something most of us probably heard of without listening: 59% of companies surveyed admit they exaggerate AI’s role in layoff announcements because, in their words, it “plays better” with stakeholders. Not because it is accurate. Because it is a more compelling story for investors than “we overhired during a period of cheap capital and now we are correcting.”

That is a significant admission. More than half of the companies publicly attributing cuts to AI are, by their own account, overstating the case.

The same Forrester report predicts that half of these AI-attributed layoffs will result in companies quietly rehiring, either offshore or at significantly lower salaries, once they discover that the promised AI capabilities are not yet ready to replace what they cut. Klarna is already a data point: the company replaced 700 customer service employees with AI, watched quality decline, faced a customer backlash, and had to bring humans back. Amazon’s “AI-powered” cashierless stores, marketed as a technological breakthrough, turned out to rely on teams of people in India watching camera footage.

The gap between what is announced and what is actually happening is wide enough to drive a very large narrative through.

I want to be precise about what I am and am not saying here, because this is not a simple story.

AI is real. Its capabilities are advancing. Some of the layoffs happening now are genuinely connected to productivity gains from AI tools, and that trend will continue. I already wrote about this and I stand by it: the differentiator will not be between humans and machines, but between people who know how to use these tools and people who don’t.

But there is a meaningful difference between a company genuinely restructuring around new capabilities and a company using an attractive narrative to justify decisions that were already being made for other reasons. Investors respond well to AI-driven efficiency stories. Boards approve restructuring packages more easily when the explanation involves the future rather than the past. The language of transformation provides cover for corrections that have nothing to do with technology.

This matters for leaders because it shapes the environment in which you are making decisions right now. If you are a senior leader reading the news and concluding that you need to reduce your engineering headcount to keep pace with industry trends, it is worth asking whether the trend you are responding to is a real structural shift or a narrative that is moving faster than the underlying reality.

Forrester’s research found that only 16% of individual workers had high AI readiness in 2025, with that number expected to reach just 25% in 2026, largely because organizations are not investing in training. HR Executive Companies are cutting the people who would have developed those capabilities, while simultaneously citing the capabilities as the reason for the cuts. That is not a transformation strategy. That is a contradiction.

IBM’s chief human resources officer put it plainly in a recent statement: the company would triple its hiring of junior talent, specifically because without entry-level engineers today, there will be no middle managers or senior engineers in ten years. The pipeline does not sustain itself.

The real story underneath the noise is this: we are in a period where the narrative around AI is running significantly ahead of the organizational reality. An NBER study across nearly 6,000 executives found that over 80% of companies report no discernible impact from AI on either employment or productivity, despite rising adoption rates. The announcements are bold. The results, so far, are mixed at best.

That gap, between story and reality, is exactly where leaders tend to make their worst decisions. Acting on a consensus that has not been verified. Moving fast on a trend that is still being understood. Cutting something that cannot be easily rebuilt.

The companies that will navigate this period well are not the ones that move fastest to align with the dominant narrative. They are the ones that have learned to ask what is actually true beneath the noise, and to build accordingly.

That, I would argue, is a more useful definition of digital leadership than anything currently appearing in quarterly earnings calls.

If this landed, share it with someone who is reading the AI headlines and trying to figure out what to actually do. They might find it useful to have a different frame.

Share