DeepMind interpretability researcher's team pivots fully to frontier AI safety, hiring in Seattle

_beenkim · x · 2026-11-19

A Google DeepMind researcher (beenkim) announces her team completely pivoted to safety about two months ago. Interpretability remains dear to her, but it was always a means to an end — she says she can't justify working on anything other than directly improving frontier model safety right now.

The team is hiring full-time roles in Seattle to work on urgent safety problems directly on Gemini, which she calls critically important 'in human history.' Application link included.

Original post →

More from Companies & People

Companies & People channel →