Former OpenAI safety lead Jan Leike joins rival company Anthropic to focus on AI safety.

From Silicon Republic: 2024-05-29 03:49:44

Former OpenAI executive Jan Leike has joined rival company Anthropic to continue the superalignment mission. Leike and former OpenAI chief scientist Ilya Sutskever resigned, leading to the dissolution of the superalignment team. OpenAI integrated superalignment across research. Leike criticized priorities, leading to his departure. Anthropic plans to expand the team to focus on AI safety.

In a recent podcast, former OpenAI board member Helen Toner explained why the board tried to fire CEO Sam Altman. Altman was briefly fired before being reinstated with a new board. Toner claimed the board lost trust in Altman due to lack of communication and inaccurate information about OpenAI’s safety processes.



Read more at Silicon Republic: Former OpenAI safety lead joins rival firm Anthropic