Johannes Heidecke, OpenAI’s head of safety systems, is leaving the company, according to an internal announcement obtained by WIRED. His exit follows a restructuring that merged the safety and research divisions, a move aimed at tightening coordination as the pace of model development accelerates.

In a staff memo, chief research officer Mark Chen said safety teams will now report to Mia Glaese, who expands her role to VP of research and safety. Saachi Jain, a former safety team lead, steps in as interim head of safety systems under Glaese. “The demands on safety continue to increase—we are training models at a much faster cadence, and release cycles have come down greatly,” Chen wrote, adding that coordination challenges around safety are now greater than ever. Heidecke, who joined in 2021 as a safety analyst and took over the safety systems role in 2024, is the latest in a string of safety-focused leaders to depart. Earlier this week, chief futurist Joshua Achiam announced he would leave after nine years.

The shakeup comes as OpenAI launched GPT-5.6, its most advanced model yet for agentic coding tasks, but flagged that it exhibited concerning misaligned behaviors compared to earlier versions. Separately, Fidji Simo, CEO of AGI deployment, is stepping down after medical leave, with Greg Brockman absorbing go-to-market strategy on top of his product oversight. The changes underscore the tension between rapid deployment and safety assurance at a company racing to push frontier models into production.

Source: Wired