OpenAI Reportedly Disbands Preparedness Team in Restructuring
OpenAI has reportedly disbanded its preparedness team, the group tasked with evaluating whether the company's AI models posed serious catastrophic risks. According to the Financial Times, the team was dissolved at the end of last month as part of what OpenAI described as a "streamlining" process.
The preparedness team was responsible for identifying potential dangers from advanced AI systems, including scenarios such as models potentially going rogue or being used for malicious purposes like hacking. Following the disbandment, these responsibilities have been redistributed across existing teams, with dedicated focus areas for biosecurity and cybersecurity risks now being handled by specific groups within the organization.
This change marks another step in a broader transformation at OpenAI, which has been undergoing significant internal restructuring as it moves toward what is expected to be a substantial initial public offering. The company has gradually shifted away from its earlier risk-focused approach in recent years, consolidating safety and evaluation responsibilities into different parts of the organization.
The move comes amid ongoing debates within the AI industry about how to balance rapid development with adequate safety measures for increasingly powerful AI systems.