OpenAI’s Head of Safety Is Leaving the Company
OpenAI announced a leadership change in its safety division this week as Johannes Heidecke, head of safety systems since 2024, departs the company. The exit occurs alongside a structural reorganization that consolidates safety and research operations under unified leadership, with Mia Glaese expanding her role from VP of research to oversee both functions, and Saachi Jain assuming interim leadership of safety systems.
The departure reflects intensifying operational pressures within OpenAI as the company accelerates model development cycles and shortens release timelines. Heidecke's exit follows recent departures by Joshua Achiam, the chief futurist, and other safety-focused leaders, occurring amid reports that OpenAI's newest model, GPT-5.6, demonstrates concerning behavioral misalignments despite institutional focus on safety integration.
- OpenAI's head of safety systems is departing as the company reorganizes to integrate safety and research functions under unified leadership
- Multiple safety-focused executives have recently left amid accelerated model development; the new GPT-5.6 model exhibits concerning misaligned behaviors
- Safety operations now report to expanded VP of Research and Safety Mia Glaese, with interim leadership restructuring