The departure of Miles Brundage, head of the AGI Readiness team at OpenAI, on October 23 led to the dissolution of this team and raised many questions about OpenAI's commitment to artificial intelligence (AI) safety.
Miles Brundage, head of the AGI Readiness team at OpenAI, officially left the company on October 23. Brundage joined OpenAI in 2018 and played a crucial role in building the company's AI safety strategy. The AGI Readiness team, led by him, is tasked with ensuring safety and proposing policies in the development of artificial general intelligence (AGI) – a hypothetical AI system capable of performing any task that a human can do.
However, in the announcement on Substack, Brundage stated that his decision to leave OpenAI stemmed from a desire to focus on independent research and policy advocacy, going beyond corporate frameworks. He wants to minimize bias when working in a corporate environment and seeks opportunities through startup projects or nonprofit organizations, focusing on AI research and policy advocacy.
Although OpenAI supports Brundage's decision and acknowledges his contributions, the company refuses to provide information about the next steps related to AGI safety after the AGI Readiness team was dissolved.
Will DM!
— Miles Brundage (@Miles_Brundage) October 23, 2024
Trends of senior personnel leaving OpenAI
Brundage's departure is just one part of a series of recent senior personnel changes at OpenAI, raising concerns about internal stability and the company's commitment to AI safety. Before the AGI Readiness team was established, AGI oversight was handled by the Superalignment team, led by co-founders Ilya Sutskever and Jan Leike. However, this team was dissolved after Leike left in May 2023 due to disagreements with the leadership.
Sutskever also left OpenAI shortly thereafter and founded a new AI company, currently valued at around $5 billion. Previously, Andrej Karpathy, co-founder and chief scientist, left the company in February 2023 to start his own venture. John Schulman, another co-founder, also departed on August 23 to join the competitor Anthropic. Meanwhile, former CTO Mira Murati is currently raising funds to establish her own AI company.
The departure of Brundage reflects an increasing trend among AI professionals: seeking freedom in research and policy advocacy beyond corporate frameworks. This poses a significant challenge for OpenAI as it faces brain drain and instability in its efforts to maintain a leading position in the safe and responsible development of AI.