An AI Researcher Just Quit, Warning AI Could Kill Us All

An AI researcher named Jacob Coxon quit his job at Anthropic this week and posted his reasons on X (formerly Twitter). His message: the big AI companies — mainly OpenAI and Anthropic — are racing to build more powerful systems without enough caution. The line that exploded online was his claim that most people building AI honestly believe it could wipe out humanity by the end of the decade. That's not a random outsider talking. That's an insider.

What made it worse is that a senior safety executive at Anthropic jumped into the thread and said, essentially, yes, that's how we feel too. If that sounds strange, consider Anthropic's pitch from day one: this technology is so powerful it could destroy humanity, so only we can be trusted to build it. It's a little like a fire company that keeps warning you your house might burn down — while selling you matches. WIRED's Will Knight says this view isn't new, but hearing it from inside has shaken people.

The episode also covered Apple's first foldable iPhone, which will cost around $2,000 and is being called the iPhone Duo. Apple's gamble is that people will pay premium prices for a phone that folds. And new "always listening" features on the Apple Watch left one host uneasy about privacy. Separately, a WIRED investigation found the US Census Bureau used faulty data to produce a misleading report about noncitizen voting that the Trump administration promoted — a reminder that bad data can shape real policy.

The takeaway isn't that AI will definitely kill us all. It's that the smartest people inside the field are worried, publicly, and still shipping products. That tension — fear and profit, side by side — is the story to watch.