OpenAI’s safety committee becomes independent with Sam Altman’s departure

OpenAI announced that the safety and security committee will become independent to prevent models from being released due to safety concerns. The move comes with a big change: CEO Sam Altman is no longer part of the safety committee.

According to OpenAI’s official website, Zico Coulter, director of machine learning at Carnegie Mellon University, will chair OpenAI’s new Safety and Security Committee (SSC). Other key members of the committee include Quora CEO Adam D’Angelo, retired US Army General Paul Nakason, and Sony General Counsel Nicole Seligman.

This new committee replaces the previous Safety and Security Committee that was formed in June 2024 and Sam Altman was a member. The former committee was tasked with making recommendations on critical safety and security decisions for OpenAI projects and operations.

Duties of the OpenAI Company Safety Committee

SSC’s responsibilities will now go beyond advising. This committee will have the authority to oversee the safety evaluations of the original models and the release of the models. Most importantly, the committee has the power to delay the release of new AI models until safety concerns are adequately addressed.

This restructuring follows a period of scrutiny regarding OpenAI’s commitment to maintaining the safety of artificial intelligence. The company has faced criticism in the past for disbanding the Superalignment team and leaving key safety-focused personnel. Altman’s removal from the safety committee appears to be an attempt to address concerns about potential conflicts of interest in the company’s safety oversight.

Other tasks of OpenAI’s security committee include plans to strengthen security measures, increase transparency about work and cooperation with external organizations. The company previously agreed to collaborate with the US and UK AI Safety Institutes on emerging AI safety risks and standards.