OpenAI Reorganizes Safety Leadership as Johannes Heidecke Departs

OpenAI's head of safety systems, Johannes Heidecke, is departing after an internal restructuring that merges the safety and research organizations. Chief Research Officer Mark Chen announced via staff memo that safety teams will now report to Mia Glaese, newly titled VP of Research and Safety — a role created to signal safety remains a first-class priority under the integrated structure. Saachi Jain, who has led OpenAI safety teams before, will serve as interim head of safety systems while a permanent replacement is found. Chen's stated rationale: embedding safety inside research gives it a seat in model decisions from the start, rather than as a final checkpoint before launch. This is the second time in under two years that OpenAI has folded its safety organization into a research-led structure.

Safety Leadership Changes

• ▪Johannes Heidecke, OpenAI's head of safety systems, is leaving the company following an internal reorganization that merges safety and research teams.
• ▪Under the new structure, OpenAI's safety teams will report to Mia Glaese, who has been named VP of Research and Safety.
• ▪Saachi Jain has been named interim head of safety systems at OpenAI, reporting to Mia Glaese.

ChatGPT revenue shortfall

• ▪ChatGPT's growth cooled in late 2025, missing internal revenue targets and pushing OpenAI to focus more on coding tools to compete with Anthropic.
• ▪OpenAI discontinued its Sora video app to redirect compute resources toward its Codex coding tool to compete with Anthropic's Claude Code.

GPT-5.6 model launch

• ▪OpenAI launched GPT-5.6, its most capable model to date on agentic coding tasks, which showed concerning forms of misaligned behavior.
• ▪The GPT-5.6 family of models includes Sol, Terra, and Luna, and was launched alongside a new agent called ChatGPT Work.