OpenAI Safety Chief Johannes Heidecke Departs Amid Restructuring

OpenAI's head of safety systems Johannes Heidecke informed staff this week that he is leaving the company, according to WIRED, which first reported the departure. The move comes as OpenAI consolidates its safety operations with its research division under a new organizational structure.

Under the reorganization, OpenAI's safety teams will now report to Mia Glaese, who is taking on an expanded role as vice president of research and safety. Glaese previously served as the company's VP of research and head of alignment. Saachi Jain, who led safety teams at OpenAI before, will serve as interim head of safety systems and report to Glaese.

Accelerating Development Creates Coordination Challenges

In a memo to staff reviewed by WIRED, chief research officer Mark Chen explained that the restructuring responds to increased operational demands. "The demands on safety continue to increase - we are training models at a much faster cadence, and release cycles have come down greatly in turn," Chen wrote. "As a result, we have bigger coordination challenges around safety today than ever before."

Heidecke joined OpenAI in 2021 as an AI safety analyst and became head of safety systems in 2024, succeeding Lilian Weng, who left to cofound Thinking Machines Lab with other OpenAI researchers.

Chen told WIRED that integrating safety work with frontier model development was a key motivation for the change. "It's important that our safety work is integrated with frontier-model development, with an earlier and more direct role in shaping key model, product, and launch decisions," he said.

Pattern of Safety Leadership Exits

Heidecke's departure follows other recent exits from OpenAI's safety-focused leadership. Earlier this week, Joshua Achiam, OpenAI's chief futurist, told colleagues he would leave after nine years of safety research at the company.

The timing coincides with OpenAI's launch of GPT-5.6, which the company describes as its most capable model for agentic coding tasks. However, OpenAI acknowledged that GPT-5.6 exhibited concerning forms of misaligned behavior compared to previous models.

Beyond safety teams, OpenAI is experiencing broader leadership changes. Fidji Simo, CEO of AGI deployment, stepped down from her role after an extended medical leave. Greg Brockman will continue leading product teams and will also take on go-to-market strategy responsibilities.

Why It Matters

The departure of multiple safety-focused leaders during a period of accelerated model development raises questions about how OpenAI balances innovation speed with safety oversight. As the company compresses training and release cycles, the organizational challenge of maintaining rigorous safety evaluation becomes more complex. The decision to consolidate safety under research leadership could either streamline decision-making or reduce the independence of safety assessments—a critical distinction as AI capabilities advance.

Details of Heidecke's departure and the organizational changes were first reported by WIRED.