Miles Brundage, a long-time policy researcher at OpenAI and senior adviser on the company's AGI readiness team, has announced his departure. He plans to shift his focus to the nonprofit sector, where he believes he can have greater influence as a researcher and advocate, with more freedom to publish his work.
In a post shared on X (formerly Twitter) and in his newsletter, Brundage acknowledged the significance of working at OpenAI, emphasizing the importance of rigorous decision-making around the development and deployment of artificial intelligence. However, he expressed that his new role will offer broader opportunities to make an impact.
Following Brundage’s exit, OpenAI's economic research division, previously part of the AGI readiness team, will be integrated under the leadership of Ronnie Chatterji, OpenAI’s new chief economist. The remaining AGI readiness projects will be distributed across other teams within the organization, with Joshua Achiam, head of mission alignment, assuming some of the responsibilities.
OpenAI expressed its support for Brundage’s decision, recognizing his contributions and commitment to AI policy research. The company expressed confidence that he would continue to influence policy discussions in his new role. However, they did not disclose who would take over his specific duties.
Brundage joined OpenAI in 2018 and led several key initiatives, including responsible deployment practices for language models like ChatGPT, the external red teaming program, and the development of OpenAI’s first "system card" reports, which document the capabilities and limitations of AI models.
In recent months, OpenAI has seen several high-profile departures, including executives like CTO Mira Murati, research VP Barret Zoph, and former chief scientist Ilya Sutskever, reflecting internal disagreements over the company’s direction. Brundage, in his post, encouraged current employees to speak up and question decisions to avoid groupthink, acknowledging the difficult choices the company faces moving forward.
The departure of Brundage follows a New York Times profile of former OpenAI researcher Suchir Balaji, who left the organization citing ethical concerns. Balaji criticized the company's use of copyrighted data in training its models, a claim also made in ongoing lawsuits against OpenAI.
This wave of leadership changes signals a period of uncertainty at OpenAI, raising questions about the future direction of the company amidst growing scrutiny.