AI is changing how you get—and keep—your job

TL;DR: Several companies this week have shown how quickly AI is upending everything about our jobs. It’s increasingly the bot you have to impress to land one, a tool you have to use in interviews—and the metric used to judge whether you deserve to keep your role.

What happened: Congrats, you (already) got the job. Now just “re-interview” for it. That’s the new hoop that privacy-focused email provider Proton is asking some staff to jump through—all to ensure they’re sufficiently AI-pilled. This interview do-over could lead to role changes and even affect “continued employment,” per a company email (though, in a Reddit comment, the firm claimed it’s “not doing layoffs and using AI as an excuse”).

While Proton is the only company we know of that’s making staff do a second audition for their jobs, it’s far from alone in using AI as a litmus test for grading employees. A larger shift is underway to sort the workers bosses think will thrive in the AI era from the ones they think will flounder.

How AI grades you: Plenty of big firms now tie performance reviews to how well employees use AI or make using it a condition of keeping the role—from Big Tech names like Microsoft and Meta to less obvious AI converts like Shopify (where using the tech “is now a fundamental expectation of everyone”) and Coinbase (which fired engineers who didn’t quickly adopt AI coding tools last year).

How AI hires you: AI also plays a bigger role in whether job seekers get a foot in the door—partly by killing off the old filters. On at least one team at Anthropic, prospective hires no longer do the dreaded coding interview, because Claude does well enough on those puzzles that they don’t predict much about candidates anymore, per a manager at the AI lab. (And since a chatbot can produce a passable cover letter in seconds, some employers have even stopped asking for one.)

But in place of these old hurdles, candidates increasingly face…more AI. Google now lets software engineering candidates use Gemini during interviews, which tests how well they prompt, according to internal documents. And then there are all the job screens now conducted by uncanny AI avatars (just don’t piss them off)—and at McKinsey, would-be consultants have to impress a fake AI client as part of the hiring process.

Bottom line: These AI yardsticks assume that the tech makes workers better and more productive at their jobs—even though the evidence for that hasn’t quite arrived yet. —WK