Former OpenAI exec that quit for ‘safety concerns’ joins rival company

Former OpenAI executive Jan Leike announced on Tuesday that he has joined Anthropic, OpenAI’s Amazon-backed rival. Leike resigned from OpenAI less than two weeks ago, publicly criticizing the company for prioritizing “shiny products” over essential “safety culture and processes.”

Leike posted on X, expressing his enthusiasm for his new role at Anthropic. Interestingly, he also urged interested readers to join him and his team.

Tweet may have been deleted

Jan Leike and others leave OpenAI

Leike’s move follows a series of notable departures from OpenAI. Just last week, policy researcher Gretchen Krueger announced her resignation, sharing similar concerns about the company’s direction.

Tweet may have been deleted

In response to these exits, Microsoft-backed OpenAI has established a safety committee overseen by board members Bret Taylor, Sam Altman, Nicole Seligman, and Adam D’Angelo. The company also revealed plans to test its “next frontier model.”

Leike co-led the Superalignment team at OpenAI with co-founder Ilya Sutskever. This team was responsible for ensuring that superintelligent AI systems remain controllable and safe. The team was disbanded after the resignations — and its members were reassigned to core research roles.

Anthropic, established in 2021 by former OpenAI employees and execs, recently secured a $4 billion investment from Amazon. Leike’s new employer has positioned itself as more safety-focused compared to its competitors — and uses Amazon Web Services’ cloud infrastructure for developing its large language models.