OpenAI has reportedly disbanded its preparedness team — the group responsible for assessing potential catastrophic risks tied to its AI models. The move comes as part of a restructuring effort described as a "streamlining" process.
The team, known as "Preparedness," was established to evaluate and mitigate serious risks posed by artificial intelligence. According to The Verge, the disbanding is part of a broader reorganization at the company.
What Happened to OpenAI's Preparedness Team
The closure reportedly took place at the end of July, with responsibilities for specific areas of preparedness work being reassigned. As MLQ.ai reports, work on biological and cyber risks has been moved to existing groups within the company.
The team's core mission was to assess the potential for catastrophic risks with OpenAI's models. This included evaluating dangers that could arise from advanced AI systems, particularly in areas like biosecurity and cybersecurity.
Safety Work Moves to Existing Groups
Rather than eliminating the safety functions entirely, OpenAI has reportedly distributed the responsibilities to other teams. According to GIGAZINE, the crisis management team that was tasked with assessing and mitigating serious AI risks has been disbanded, but the underlying work continues in different parts of the organization.
"OpenAI quietly disbanded its preparedness team at the end of last month, with responsibilities for individual areas of the preparedness work being reassigned." — Theo Bearman via X
The restructuring appears to be part of a larger streamlining effort at OpenAI. The company has not publicly detailed the reasons for the change, but the move signals a shift in how the organization approaches AI safety oversight.
What This Means for AI Safety Oversight
The disbanding raises questions about how OpenAI will maintain independent risk assessment as it continues developing increasingly powerful models. The preparedness team was specifically designed to evaluate catastrophic risks — worst-case scenarios that could have severe consequences if left unchecked.
Key points to understand about this development:
- The preparedness team was a dedicated unit focused on catastrophic risk assessment
- Its work on biological and cyber risks has been reassigned to existing groups
- The disbanding is part of a "streamlining" process at OpenAI
- The closure happened at the end of July, according to reports
Our Take: A Questionable Streamlining Decision
To put it plainly, this decision deserves scrutiny. OpenAI is one of the most influential AI companies in the world, and its models are becoming more capable by the day. Disbanding a dedicated team that focuses specifically on catastrophic risks — even if the work is reassigned — raises legitimate concerns.
When safety responsibilities are spread across existing groups, they can easily become secondary priorities. A dedicated team has clear focus and accountability. That focus gets diluted when risk assessment becomes just one more task among many.
The "streamlining" rationale is understandable from a business perspective, but AI safety is not an area where efficiency should come at the cost of thorough oversight. We believe OpenAI should be transparent about how the reassigned work will be monitored and who will be accountable for catastrophic risk assessment going forward.
This is a story worth watching closely. How OpenAI handles this transition will set a precedent for how other AI companies approach safety infrastructure during periods of growth and restructuring.
