OpenAI is losing another safety leader as Johannes Heidecke, the company's Head of Safety, exits amid a restructuring plan to integrate its research and safety operations more tightly. The departure, first reported by Wired, adds to a growing list of safety-focused executives who have left the company behind ChatGPT, raising questions about how the AI lab balances innovation with responsible development. OpenAI faces mounting pressure from regulators, researchers, and its own employees about moving too fast with increasingly powerful AI systems. The company says the restructuring is about efficiency and better collaboration, aiming to have safety experts work side-by-side with researchers rather than as separate oversight teams. Critics argue integrating safety into research teams could blur independent oversight. Heidecke's exit follows other high-profile departures from OpenAI's safety ranks, including Jan Leike, who co-led the superalignment team and left in May 2024 with pointed criticism, and Ilya Sutskever, co-founder and chief scientist, who departed around the same time. The company insists it remains committed to responsible AI development.