A researcher walked out

Jacob Coxon resigned from Anthropic last week over safety concerns. He had worked as a safety researcher at both Anthropic and OpenAI.

His exit post went wide

On X, Coxon wrote that neither lab is acting responsibly, and the post spread far past what he expected.

“They are racing straight to self-improving superintelligence and gambling with our lives.”

The fear inside the labs

Coxon says the worry is not hypothetical for the people building these systems. He put their timeline at the end of the decade.

“The people building AI earnestly believe that it could kill us all by the end of the decade.”

Then Amodei blinked

On Saturday, Anthropic CEO Dario Amodei published a letter titled "We Must Pace the Frontier," calling on AI companies to slow down.

“I believe we owe it to humanity to try.”

Coxon calls it a start

Speaking to George Stephanopoulos on "This Week," Coxon welcomed the letter as movement toward the coordinated slowdown he wanted.

“I think this is a substantial step towards what I was talking about.”

Risk talk is now routine

Coxon says big-risk conversations now come up a couple of times a week across the company, no matter where you sit in it.

“There's not enough time to be constantly thinking about the risks.”