OpenAI announced an internal restructuring plan this Tuesday. According to a post by the company's CEO, Sam Altman, on the social media platform X, Aleksander Madry, the former head of AI safety, has been reassigned to a new role focused on AI reasoning.
Altman praised Madry's work in a tweet, stating that he is "working on a new, very important research project." This move not only acknowledges Madry's capabilities but also underscores OpenAI's commitment to research in the field of AI reasoning.
During Madry's reassignment, OpenAI executives Joaquin Quinonero Candela and Lilian Weng will jointly take over the Preparedness team. This team is primarily responsible for identifying new risks in cutting-edge models and conducting safety assessments to mitigate potential "catastrophic risks" posed by AI.
It is noteworthy that although Madry's focus has shifted, OpenAI has stated that he will continue to be involved in core AI safety work. Madry holds key positions at OpenAI and is also the director of the Deployable Machine Learning Center at MIT, as well as a faculty co-director of the MIT AI Policy Forum. His extensive experience and expertise are expected to bring new breakthroughs to OpenAI's AI reasoning research.
This restructuring reflects OpenAI's efforts to simultaneously maintain AI safety research and actively advance the development of AI reasoning capabilities. Industry insiders generally believe that this strategic adjustment will help OpenAI maintain its leading position in the fiercely competitive AI field.