OpenAI has announced significant changes to its safety and security practices, including the creation of a new independent board oversight committee. The move comes with a notable change: CEO Sam Altman will no longer be a member of the safety committee, marking a departure from the previous structure.
The newly established Safety and Security Committee (SSC) is chaired by Zico Colter, Dean of Machine Learning at Carnegie Mellon University, and other key members include Quora CEO Adam D’Angelo, retired U.S. Army General Paul Nakasone, and Nicole Seligman, former EVP and General Counsel at Sony Corporation.
The new committee replaces the previous Safety and Security Committee, which was formed in June 2024 and included Altman as a member. The original committee was tasked with making recommendations on key safety and security decisions regarding OpenAI’s projects and operations.
The SSC’s responsibilities go beyond making recommendations: it will oversee safety evaluations of major model launches and will have the power to monitor model launches. Importantly, the committee will have the power to delay a launch until safety concerns are adequately addressed.
The restructuring comes after a period of scrutiny over OpenAI’s AI safety efforts; the company has previously faced criticism for disbanding its superalignment team and the departure of key safety-focused personnel. Removing Altman from the safety committee appears to be an attempt to address concerns about potential conflicts of interest in the company’s safety oversight.
OpenAI’s latest safety measures also include plans to strengthen its security measures, increase transparency about its efforts, and collaborate with external organizations. The company has already reached agreements with AI safety research institutes in the U.S. and the U.K. to collaborate on research into emerging AI safety risks and standards for trustworthy AI.
