OpenAI's head of safety is leaving the company as part of a reorganization that eliminates the standalone role.
The departure is not a simple resignation — it reflects a structural decision to fold safety oversight into a broader executive position covering both research and safety teams. The dedicated head of safety role will not be backfilled. Instead, a single executive will now hold authority over both functions, consolidating two teams that have historically operated with some independence.
That consolidation is the part worth watching. Safety and research teams at AI labs are often in productive tension — safety work exists partly to scrutinize and slow down research outputs. Merging them under one leader removes an organizational check. Whether the new structure strengthens safety by giving it more resources and authority, or dilutes it by subordinating it to research priorities, depends entirely on who takes the combined role and how they run it.
OpenAI has lost several high-profile safety figures in recent years, a pattern that has drawn repeated criticism from researchers and policymakers. Absorbing the safety function into research leadership is a choice that will be read, fairly or not, as a signal about where safety ranks in the company's priorities.