According to Financial Times, OpenAI dissolved its safety preparedness team late last month. The team was responsible for assessing whether models presented serious risks and developing ways to mitigate those risks.
According to FT, the responsibilities were redistributed to specific areas such as biosecurity and cybersecurity, then integrated into existing teams. The change represents the latest reorganisation at the company, which has been undergoing transformation as it prepares for what is expected to be a major public listing.
In recent years, the company has gradually dismantled its responsible research model, dissolving key departments.