"I lost trust": Why the OpenAI team in charge of safeguarding humanity imploded

Several safety-focused OpenAI employees have left the company, and the article attributes this mainly to lost trust in CEO Sam Altman rather than to a technical breakthrough. Ilya Sutskever and Jan Leike announced their departures from the superalignment team, and Daniel Kokotajlo, who worked on the governance team, quit last month, saying he had gradually lost trust in OpenAI leadership's ability to handle AGI responsibly. The article also cites the firing of Leopold Aschenbrenner and Pavel Izmailov, whom media reports said were fired for leaking information, though OpenAI has offered no evidence of a leak. OpenAI has since replaced Leike with John Schulman, and the article says the superalignment team had been promised 20 percent of the company's computing power.