Joe Benton AI safety researcher focused on evaluating advanced systems and alignment. He previously worked on alignment at Anthropic and completed a PhD at the University of Oxford.
Joe Benton AI safety researcher focused on evaluating advanced systems and alignment. He previously worked on alignment at Anthropic and completed a PhD at the University of Oxford.
Career
AI Safety Researcher at Anthropic until 2026-09Cosigns Received
First Cosign
Joe is among the most talented and ambitious AI safety researchers in the world, and his work was a big reason why i saw Anthropic as a place where good technical AI safety work could actually happen seeing him leave is a very interesting signal