Will AI wipe out half of entry-level jobs? What the AI bosses said, and what's happened since

What he actually said

In an interview with Axios in May 2025, Amodei said AI could wipe out half of all entry-level white-collar jobs and push unemployment to 10 to 20% within one to five years. He named the fields he saw as most exposed at the start of a career: tech, finance, law and consulting. His wider point was that the companies building AI, his own included, should stop softening the message.

You may have seen a version going round that has him listing an entry-level lawyer, a financial analyst and a junior consultant, with the advice to learn to use AI. We couldn't find that exact wording in a transcript, so we aren't quoting it. The idea is the same one he gave Axios.

He isn't the only one saying it

In June 2025, Amazon CEO Andy Jassy wrote to staff that AI "will reduce our total corporate workforce as we get efficiency gains from using AI extensively across the company." In the same memo he wrote: "We will need fewer people doing some of the jobs that are being done today, and more people doing other types of jobs."

Fewer people doing some of today's jobs, more doing other kinds. That second half is where your next move is.

What the data shows now

• The first rung is narrowing. Stanford's Digital Economy Lab, using ADP payroll data, finds employment for 22 to 25 year olds in highly AI-exposed jobs is 19% below where it would be had it kept pace with less-exposed peers.
• Experienced workers in the same jobs show no comparable gap.
• It's fewer hires, not mass firings. The gap comes mainly from companies hiring fewer young people.
• It depends how AI is used. Declines cluster where AI does the task instead of a person. Where AI helps a person, employment is flat or rising.
• The whole market hasn't broken. Yale's Budget Lab and Brookings found no discernible disruption to the wider US labour market in the 33 months after ChatGPT launched.

So both things are true. Most jobs are still there. The jobs that used to be how people learned are getting scarcer, and the researchers call their figures early warning signs rather than proof of cause.

Why the first rung goes first

Look at what juniors were hired to do: first drafts, research summaries, cleaning data, basic code, slide decks and document review. That's the work AI is now quickest at. A senior person with AI can do more of it themselves, so a team that once hired three juniors may hire one.

What the senior person still can't hand off is judgement: spotting the draft that's wrong, hearing what the client really meant, noticing the number that doesn't add up. That's where entry-level value is moving.

What to do about it

1. 01Use AI on real work every week: your coursework, a side project or your job, not just tutorials.
2. 02Check, don't just generate. Practise catching AI's mistakes, because that's the skill employers can't find.
3. 03Finish things. A finished project you can explain beats a certificate.
4. 04Aim where AI helps people. Roles that mix judgement, people and hands-on work are where employment is holding up.

Test a future job before you train for it

A forecast can tell you which jobs are growing. It can't tell you whether you'd enjoy the work. Laddie, our AI coach, starts from the experience you already have and works through four steps with you:

1. 01Explore: directions your experience carries into, including work that's only now taking shape as AI changes jobs.
2. 02Try: an experiment small enough to finish this week, a small piece of the work itself.
3. 03Learn: note which part you liked and which part you avoided, and keep it as a record of what you tried.
4. 04Apply: when a direction turns into a job post, OffLadder CV shapes your CV for it. What you can't show yet goes back to Laddie as a short lesson, a 15-minute task and a mock interview.