Here's a story you've probably heard: artificial intelligence is coming for our jobs. From ServiceNow, Inc. (NOW) to Block, Inc. (XYZ), and even from influential voices like Larry Fink, the narrative is building that AI is driving layoffs. Fink has warned that AI could push unemployment higher, especially among younger, white-collar workers.
It's a clean story. It's also, well, a pretty convenient one.
Because when you actually look at the numbers, the picture gets a lot murkier.
The Data Isn't Screaming ‘AI Disruption'
Think about it: if AI were already displacing workers on a large scale, you'd expect to see it first in higher-skilled roles—the exact jobs most vulnerable to automation. That's the logical place for the disruption to show up.
But that's not what's happening.
Recent data shows unemployment among college graduates is still meaningfully lower than for workers without degrees. That gap is intact and only narrowing gradually—it's not collapsing in a way that would signal a sudden AI shock. Separate research and analyst commentary have also pointed out that AI hasn't yet made a material contribution to job losses at the macroeconomic level.
In other words, the labor market still looks cyclical. It doesn't look like it's been structurally disrupted by a new technology. At least, not yet.
Blaming AI, Fixing Balance Sheets
So, if the data isn't showing a massive AI-driven job apocalypse, why is AI getting so much of the credit—or blame—for layoffs?
A big part of the answer comes down to incentives.
Framing job cuts as "AI-driven" does a few clever things for a company. It signals efficiency and technological savvy to investors. It can help avoid some of the political and public relations scrutiny that comes with cutting domestic jobs. And perhaps most importantly, it aligns the company with a forward-looking, innovative narrative instead of a backward-looking story about correcting past mistakes.
Take Block, for example. The company has cut a significant chunk of its workforce, with some reports putting the figure as high as 40–50%. While automation and efficiency have been cited, analysts and industry voices have pointed to COVID-19 pandemic-era overhiring as a key driver. They're not alone. Executives like Salesforce, Inc.'s (CRM) Marc Benioff have argued that many companies expanded too rapidly during the boom times and are now course-correcting.
Reports on Block's job cuts similarly describe them as a reset after that COVID-era hiring spree, rather than a purely AI-driven transformation.
AI might be part of the story. But it's probably not the main character.
Profitability Is Improving — But Why?
Let's be clear: AI is absolutely helping with margins, at least around the edges. Automation tools, AI copilots, and workflow improvements are real and they're delivering value.
But the timeline here is crucial.
Most companies are still very early in deploying AI at any meaningful scale. Research from McKinsey shows that the majority of firms are still in the pilot or experimentation phases and haven't yet achieved enterprise-wide impact. That makes it pretty unlikely that AI alone is the prime mover behind the current wave of job cuts.
Instead, what we're probably seeing is a mix of a few different things:
- The unwind of post-pandemic overhiring
- A general return to cost discipline as the economic environment shifts
- AI providing a convenient, shiny narrative overlay for all of the above
Narrative Vs. Reality
None of this means AI won't reshape the workforce. It almost certainly will, eventually. But right now, the market and corporate messaging might be pulling that impact forward in time.
For investors, this distinction is critical. If the layoffs are truly AI-driven, then the disruption is structural—and we're just at the beginning of a much longer trend. If they're mostly driven by overhiring corrections and margin resets, then this particular cycle of job cuts might already be further along.
At the moment, executives are telling one story. The data, for now, is quietly suggesting another. It's worth listening to both.