Is AI taking jobs?

The answer is not so obvious

1. Most of the world's media quotes a dubious figure from a single website

"Challenger, Gray & Christmas" is the name of an Illinois company that sells career consulting: it literally earns its money from layoffs. This very firm began publishing reports about "mass layoffs at companies due to AI"¹.

The reports brought huge traffic: CNN, NBC, CBS News and plenty of other outlets quote them. That single source helped launch the global "AI is going to take my job" scare story.

Where does "Challenger, Gray & Christmas" get this data from in the first place? They do not disclose their methodology, but they have no access to real statistics from government bureaus.

The economist Matt Darling calculated that their report captures about 3% of all real layoffs².

If you saw a headline claiming "23% of layoffs are happening because of AI", there is a good chance it came from the same HR blog - unverified.

This year the real mass layoffs are in transport, up 387% year on year, but you have probably seen far fewer headlines about that, because Terminator stories get more clicks.

2. No mass displacement, but an entry-level warning sign

Aggregate labour data still shows no economy-wide displacement attributable to AI.

• European Central Bank analysis of around 5,300 euro area firms found no significant difference in job creation and destruction between firms that use AI and those that do not. Firms that use AI intensively were about 4% more likely to take on additional staff. That is an association in a firm survey, not evidence that AI creates jobs³.
• PricewaterhouseCoopers analysed a billion job advertisements across 27 countries and found that the wage premium for AI skills rose from 57% to 62% over the past year⁴.

The more specific warning comes from a revised Stanford Digital Economy Lab study by Erik Brynjolfsson, Bharat Chandar and Ruyu Chen. It uses high-frequency payroll data from Automatic Data Processing (ADP) covering millions of workers in the United States through June 2026⁵.

Workers aged 22 to 25 in highly AI-exposed occupations were about 19% below the employment level they would have reached if they had kept pace with similarly aged workers in less-exposed occupations. This is a relative gap against a comparison trajectory, not an absolute 19% decline in employment.

In absolute levels, employment among workers aged 22 to 25 in the two most AI-exposed occupation groups fell about 11% between November 2022 and June 2026, while in the three least-exposed groups it grew about 10%. The divergence appears mainly through reduced hiring of young workers rather than increased separations, and it is concentrated in occupations where AI tends to substitute for human tasks. Where AI complements workers, employment is flat or rising, especially among experienced workers⁵.

The Stanford authors call these descriptive patterns, not causal estimates, and the warning is narrow: entry-level jobs are under pressure in some AI-exposed occupations⁵.

The labour market is changing, just not in the way the panic headlines suggest: AI is reshaping roles inside offices.

Some firms are hiring specialists who combine humanities knowledge and AI: marketing engineers, design engineers and UX engineers.

A marketer with LLM engineering skills, who can build a dozen AI agents for the company himself, and these agents do not replace him but make him stronger.

3. It is advantageous for company management to blame layoffs on AI, and this is called AI washing

Companies do not like to admit that they are laying people off because the business is doing badly. Layoffs because of AI are another matter, since that sounds technological. "We are laying off employees not because our business is collapsing, but because we have flown off into technological outer space! Super-robots now do the work of seven people."

(People who have actually set those "robots" up themselves know that AI does not work without a human watching over it. Do not believe the fairy tales from Twitter.)

If you are laying people off without a mature, ready-to-go AI agent to do the work, you are not laying off people because of AI.⁶

Post-Covid overhiring is one major explanation for the 2025-2026 technology layoffs, while AI has also become a convenient justification for some companies.⁷

Marc Andreessen is one of the most successful businessmen in Silicon Valley. He estimates the overstaffing at up to 75% of the headcount, and he openly calls AI the silver-bullet excuse.

That does not mean every AI-related layoff is fake, only that the label needs evidence.

Klarna is the obvious case: it loudly said an AI assistant had replaced 700 people, then started hiring humans again⁸.