The AI jobs apocalypse isn't in the data. Something quieter is.

Ask a room of executives whether AI is destroying jobs and most will say yes. The data says no. Not yet, and not the way people think.

But there is a real problem sitting under the noise, and almost nobody is naming it. That gap between the fear and the facts is exactly what this newsletter exists to close. So let us check the hype.

REALITY CHECK

AI is triggering mass job loss across the economy. You have read the headline a hundred times.

The macro picture is calm. The Yale Budget Lab looked at the 33 months since ChatGPT launched and found no discernible economy-wide disruption. The mix of who works in which occupation barely moved.

A Danish study of 25,000 workers across 7,000 firms found the same thing up close. Null effects on pay and hours, ruling out anything larger than 2 percent two years after ChatGPT. Average time saved was 2.8 percent of work hours. Real, but not the stuff of collapse.

And the longer view points up, not down. The World Economic Forum projects a net gain of 78 million jobs by 2030, with 170 million created against 92 million displaced.

So far this looks like every hype cycle. Then you check who is getting hired, not just who is employed. And the story changes.

Stanford tracked payroll data from ADP and found early-career workers, ages 22 to 25, in the most AI-exposed jobs saw a 13 percent relative drop in employment. Entry-level software and customer service roles fell close to 20 percent from late 2022 to mid-2025. Anthropic ran its own numbers and found hiring of 22 to 25 year olds into exposed roles slowed about 14 percent. Older, experienced workers in the same jobs held steady or grew.

The jobs apocalypse is hype. The entry-level squeeze is real, and it is underreported. AI is not firing people at scale. It is quietly removing the bottom rung of the ladder.

The people building and banking on AI don’t agree on much. Notice where they do.

“You’re not going to lose your job to an AI. You’re going to lose your job to someone who uses AI.” Jensen Huang, CEO of NVIDIA (Milken Institute, May 2025).

“[AI] will eliminate jobs. That doesn’t mean that people won’t have other jobs.” Jamie Dimon, CEO of JPMorgan Chase (December 2025).

Anthropic’s own CEO, Dario Amodei, warned AI could wipe out half of all entry-level white-collar jobs and push unemployment to 10 to 20 percent within five years (Axios, May 2025). The floor, not the ceiling, is where the risk sits.

If you run a team, your risk is not a headcount bloodbath. It is a training gap you will feel in three to five years.

Entry-level roles are how people learn judgment. Cut them to the bone because AI can do the rote work, and you stop producing the mid-level people you will need later. You will have automated your own pipeline out of existence and not noticed until the shortage hits.

A Reuters and Ipsos poll last August found 71 percent of Americans fear AI will take their job. Most of them are worried about the wrong quarter. The people who should worry are new graduates, and the leaders who quietly depend on a steady supply of them.

Two moves worth making now. Keep some entry-level roles and redesign them around judgment and AI oversight, not the rote tasks AI already handles. And track your own hiring by age band, so you can see the ladder thinning before it turns into a gap you cannot fill.

The hype tells you to fear the wrong thing. The data tells you where to actually look.

Three things worth your attention this week.

AI. Executives are souring on their own AI rollouts.

48% now call it a disappointment, up from 34% a year ago, and 75% admit their AI strategy is more for show. Mark’s take: the gap isn’t the models, it’s deployment discipline. Pick two proven use cases and kill the vanity projects.

Cyber. Novo Nordisk, the maker of Ozempic, got hit.

Attackers reached internal systems and copied personal data, disclosed June 11. Mark’s take: steal-and-leak, not lock-and-ransom. Assume your data walks even when nothing gets encrypted.

Cyber. A perfect-10 bug is under CISA’s spotlight.

A maximum-severity authentication bypass in SimpleHelp remote-support software, CVE-2026-48558. Mark’s take: remote-support and RMM tools are still a favorite front door. Patch this one today, not this quarter.

23% of enterprises report significant ROI from their AI agents. 97% have deployed them this year.

Source: 2026 enterprise AI adoption surveys.

What AI or cybersecurity claim do you want checked next? Reply and tell me. The next Hype Check might be yours.

The Budget Lab at Yale, “Evaluating the Impact of AI on the Labor Market” (2025-2026)

Humlum and Vestergaard, “Large Language Models, Small Labor Market Effects,” NBER Working Paper 33777 (2025)

Brynjolfsson, Chandar, and Chen, “Canaries in the Coal Mine? Six Facts About the Recent Employment Effects of Artificial Intelligence,” Stanford Digital Economy Lab (2025)

Anthropic Economic Index, “Labor Market Impacts of AI” (March 2026)

World Economic Forum, “Future of Jobs Report 2025”

Reuters and Ipsos poll on AI job fears (August 2025)

Mark Lynd is a 5x CEO, CIO, and CISO, 4x book author, and keynote speaker on AI and cybersecurity. He advises C-suites and executive teams on building efficient and resilient organizations. Learn more and book him at marklynd.com. He also publishes Hype Check Now, a newsletter on signal over noise in AI and cybersecurity, read by thousands of technology and security leaders worldwide.