OpenAI's head of safety systems, Johannes Heidecke, has announced his departure from the company after taking on the role in 2024. His exit comes as OpenAI aims to launch increasingly advanced AI models, with GPT-5.6 being the most recent and controversial release.

Mark Chen, chief research officer, outlined plans for restructuring safety teams under Mia Glaese, VP of research and head of alignment. Saachi Jain will temporarily lead these efforts until a permanent successor is named.

The increasing complexity in model development poses significant coordination challenges, particularly as misaligned behaviors have surfaced in newer models. This mirrors concerns raised by Joshua Achiam, OpenAI's chief futurist, who also resigned after nine years of research.

Meanwhile, Fidji Simo, CEO of AGI deployment, has stepped down following an extended medical leave, with Greg Brockman taking on both product and go-to-market strategy roles. These changes reflect the rapid pace of development within artificial intelligence and the ongoing need for robust safety measures.