An update on OpenAI safety & security practices

OpenAI established the Safety and Security Committee to make recommendations on critical safety and security decisions as OpenAI models reach each new level of capability and continue solving hard problems for people around the world.

As one of its initial mandates, the Safety and Security Committee conducted a 90-day review of safety and security-related processes and safeguards and made recommendations to the full Board.