OpenAI Decentralizes OpenAI for Science to Empower Research Teams
- OpenAI has announced the departures of two senior executives amid a broader leadership reshuffle, with the company simultaneously decentralizing its OpenAI for Science initiative to better align with...
- Bill Peebles, who led OpenAI's short-form video application Sora, and Kevin Weil, the vice president of OpenAI for Science, both confirmed their exits on Friday, April 17, 2026,...
- An OpenAI spokesperson said the company is decentralizing OpenAI for Science in an effort to bring its work closer to the teams that are building leading model capabilities,...
OpenAI has announced the departures of two senior executives amid a broader leadership reshuffle, with the company simultaneously decentralizing its OpenAI for Science initiative to better align with core product and infrastructure teams.
Bill Peebles, who led OpenAI’s short-form video application Sora, and Kevin Weil, the vice president of OpenAI for Science, both confirmed their exits on Friday, April 17, 2026, through separate posts on X. Srinivas Narayanan, chief technology officer of B2B Applications at OpenAI, also announced his departure on the same day.
An OpenAI spokesperson said the company is decentralizing OpenAI for Science in an effort to bring its work closer to the teams that are building leading model capabilities, products and infrastructure. The move follows recent leadership changes, including Fidji Simo, OpenAI’s product and business chief, taking a medical leave due to a worsening neuroimmune condition, and Kate Rouch, the marketing chief, stepping down to focus on cancer recovery earlier in April 2026.
Kevin Weil, who joined OpenAI a couple of years ago as chief product officer after leading product teams at Twitter and Instagram, previously pursued a PhD in particle physics at Stanford University before transitioning to Silicon Valley. His leadership of OpenAI for Science was highlighted in an earlier interview with MIT Technology Review, where he discussed the team’s mission to make scientists more productive through AI-powered research tools, particularly leveraging GPT-5.
