OpenAI Disbanded Its Preparedness Team at End of Last Month, FT Reports
OpenAI has reportedly disbanded its internal preparedness team, according to the Financial Times, ending a unit tasked with judging whether frontier models pose serious risks and designing mitigations. Reporting framed the move alongside broader executive reshuffles and staff frustration as the company prepares for a potential IPO, suggesting organizational pressure toward commercial milestones. The change arrives near other governance shifts, including the reported July departure of head ethicist Chloé Bakalar without a named replacement, with descriptions of safety work shifting into engineering rather than standalone oversight. For builders and enterprise operators, the signal is not a confirmed policy reversal but a reduction in dedicated, independently staffed risk assessment capacity at a major API provider. Attribution remains to FT sourcing and secondary coverage; OpenAI has not detailed replacement structures in the packet.
OpenAI Disbanded Its Preparedness Team at End of Last Month, FT Reports
According to the Financial Times, OpenAI disbanded its preparedness team at the end of last month. The job of the preparedness team was to assess if models posed serious risks and develop ways to mitigate those risks.
Key takeaway
The disbanding of OpenAI's preparedness team removes a dedicated internal unit for frontier risk assessment, a structural change operators should weigh against public safety claims.
What happened
According to the Financial Times, OpenAI disbanded its preparedness team at the end of last month. The team had been responsible for assessing whether models posed serious risks and developing ways to mitigate those risks.
Techmeme reporting citing Financial Times sources links the move to repeated executive reshuffles and staff frustration as OpenAI prepares for a potential IPO. Separate reporting also notes head ethicist Chloé Bakalar left in July without a replacement.
Evidence
OpenAI disbanded its preparedness team at the end of last month per Financial Times reporting.
The Verge AI · attributed
According to the Financial Times, OpenAI disbanded its preparedness team at the end of last month.
The preparedness team assessed serious model risks and developed mitigation strategies.
The Verge AI · attributed
The job of the preparedness team was to assess if models posed serious risks and develop ways to mitigate those risks.
Financial Times sources tie the disbandment to executive reshuffles and IPO preparation.
Techmeme · attributed
Sources: OpenAI's repeated exec reshuffles and departures have frustrated some staff as it prepares for an IPO; it disbanded its "preparedness" team in July (Financial Times)
Head ethicist Chloé Bakalar left OpenAI in July and was not replaced.
Hacker News AI · attributed
Chloé Bakalar, OpenAI's head ethicist, left the company in July, leaving the role vacant.
Why it matters
Enterprise buyers and policymakers may need to reassess how independently OpenAI evaluates catastrophic model risks now that IPO preparation coincides with executive churn and ethics leadership vacancies.
Limits and uncertainties
OpenAI has not publicly confirmed the restructuring details or named successors in the available reporting.
Gizmodo coverage in the packet is limited and points mainly to Hacker News discussion rather than original Financial Times sourcing.
Practical implications
API customers should track OpenAI safety policy updates and model cards for signs that risk review processes moved into product teams.
Compliance teams should document which internal governance assurances depended on a standalone preparedness function.
What to watch
Whether OpenAI announces a replacement safety governance structure or publishes updated preparedness evaluations for new frontier models.
Further Financial Times or official OpenAI statements clarifying timing and staff assignments after the July disbandment.
