Lilian Weng has returned to OpenAI as VP of AI Safety, The Information reported. She will work on recursive self-improvement research—systems that help improve the processes used to build future AI systems.
The move follows Weng's July 28 departure from Thinking Machines Lab, which she co-founded in 2025. Before that, she spent nearly seven years at OpenAI, where she led applied research and the Safety Systems team.