Yale Says No Jobs Have Been Lost to AI — as Amazon Lays Off 14,000 and Business Insider Declares ‘The Era of Mega AI Layoffs Is Here’. What gives?
Executives are blaming AI for layoffs, but the data—and their own retractions—tell a different story. So what's the real story?
Amazon announced plans to lay off up to 30,000 corporate employees and recently laid off 14,000. The reason? “Transformational advancements in AI.”
Meanwhile, a new study from Yale University’s Budget Lab, Evaluating the Impact of AI on the Labor Market: Current State of Affairs, found that across the entire U.S. labor market no measurable job loss has yet been caused by AI. The authors describe the picture as “continuity rather than collapse.”
So what gives?
The story vs. the statistics
The Yale team compared employment across hundreds of “AI-exposed” occupations—data analysts, copywriters, administrative staff—and found no statistical link between exposure to AI and job loss.
"Measures of exposure, automation, and augmentation show no sign of being related to changes in employment or unemployment.”
— Yale Budget Lab, 2025
In other words, the story is moving faster than the statistics: AI is nibbling at the edges, not hollowing out the core. Most systems still need human supervision, context, and correction.
AI today is a confident intern — fast, articulate, occasionally brilliant, often wrong. You wouldn’t give it your P&L or your kids. But you might let it draft the first paragraph.
Belief as a management strategy
Still, belief moves faster than reality.
When Copilot promises to make developers 30 percent more productive, leadership hears, "cut 30 percent of developers." When a language model can write a decent blog post, they ask, "why do we still have a marketing department?"
Layoffs justified by AI are rarely about what’s true today — they’re about what leaders wish were true tomorrow. It’s the same pattern we saw during the “digital transformation” era: an idea so fashionable it started making decisions for people.
Faith in efficiency has become its own performance metric.
The illusion of inevitability
Executives love inevitability. It’s cleaner than accountability.
If a leader says “We’re cutting costs,” they’re the villain.
If they say “AI is transforming everything,” they’re a visionary adapting to the future.
That’s what’s happening here. AI has become the perfect narrative device: Symbol and scapegoat in one sentence.
It lets companies make old-fashioned cuts and call them modernization. Layoffs become “transformation.” Austerity becomes “innovation.”
We’re not automating people out of work.
We’re rebranding the act of firing them.
And automating that too.
The bold claims — and the quiet retractions
Bold headlines sell the story.
Retractions reveal the truth.
The gap between them is where the myth of AI-driven layoffs lives.
I’m not just making that up. Don’t take my word for it—take Andy Jassy, who runs a small online dropshipping business.
| Executive | Initial Claim | Later Retraction / Softening |
|---|---|---|
| Andy Jassy (Amazon) | “As we roll out more Generative AI and agents … we will need fewer people doing some of the jobs that are being done today.” (Reuters, Jun 2025) | “These cuts aren’t really AI-driven … this is about culture and focus.” (Fortune, Nov 2025) |
| Marc Benioff (Salesforce) | “I’ve reduced [support staff] from 9,000 heads to about 5,000 because I need less heads.” AI agents now handle ≈50 % of customer conversations. (Business Insider, Sep 2025) | “Look, we love AI … but AI doesn’t have a soul. It’s not that human connectivity.” Then announced plans to hire thousands of new salespeople. (Business Insider, Oct 2025) |
| Satya Nadella (Microsoft) | “If there is one sector that's getting good and more efficient, some of the labour force in that sector will disperse more broadly” (Moneycontrol News, Feb 2024) | "So, the first thing that we all have to do is, when we say this is like the Industrial Revolution, let’s have that Industrial Revolution type of growth.” (Futurism, Feb 2025) |
AI as a narrative container
AI is no longer just a technology — it’s a story we tell about change. A vessel big enough to hold every anxiety and ambition of the moment.
For executives, it justifies disruption.
For investors, it explains volatility.
For policymakers, it offers a neutral villain when things get messy.
When growth stalls, it’s AI’s fault.
When profits spike, it’s AI’s brilliance.
It’s a mirror for an age of plausible deniability — a way to say something big is happening without naming what’s really driving it.
Even Federal Reserve Chair Jerome Powell has joined the chorus:
“Job creation is pretty close to zero,” he warned, “as AI-fuelled investment props up growth but erodes hiring.”
(Fortune, Oct 2025)
It’s a striking line — but like the CEOs above, it pins the spotlight on AI while the rest of the stage goes dark. The economy is far more complicated: policy lags, capital concentration, decades of structural imbalance. AI just happens to be the most cinematic explanation available.
Efficiency theater
When stories get ahead of systems, operations follow suit.
Inside companies, the myth of AI efficiency quickly becomes the practice of AI austerity.
Every organization wants to be lean. But “lean” can easily become brittle.
When you remove too many people who understand how things actually work, you get a short-term margin boost and a long-term knowledge debt.
These are the quiet disruptions that don’t make headlines: slower responses, broken hand-offs, institutional memory evaporating.
Efficiency without purpose isn’t clarity—
it’s collapse rehearsing as progress.
The state of the technology
I’ve implemented AI solutions across multiple platforms and use them extensively.
They’re powerful — astonishing, even. They can accelerate, amplify, and occasionally delight. But they’re not magic.
I’m acutely aware of AI’s present state and its promise — and maybe even where the puck is heading. That’s why I’m skeptical.
AI can summarize, suggest, and simulate. It can compress the cost of work, but not replace the quality of judgment. It cannot yet understand.
And judgment — context, nuance, discernment — is still the thing holding every system together.
So what gives?
Yale’s data and Amazon’s layoffs aren’t contradictions — they’re two halves of the same story.
At scale, the labor market is steady. Inside the C-suite, it’s prophecy and PowerPoint about the “AI era.”
The jury is still out on how much of this is real automation and how much is convenient mythology. We’ll find out not in press releases, but in performance. The revolution will not be press-released.
If AI truly makes companies better, we’ll see the proof: faster service, smarter operations, fewer errors. If not, we’ll see the seams—dropped tickets, slower updates, messier products. (The technical term for this is “enshittification.”)
Yale will keep counting jobs. Amazon will keep counting savings. The rest of us will be counting the lag—the distance between what AI can do and what we say it can do.
Either way, the next few quarters will tell the truth.
And if those services stumble?
Well — maybe the AI that caused the layoffs can help fix them.