Openai’s Head of Safety Systems Johannes Heidecke Announces Departure
Johannes Heidecke told staff he is leaving OpenAI as the company reshapes its safety and research teams. The move follows a memo from chief research officer Mark Chen outlining a new reporting structure and an interim leadership plan.
OpenAI’s head of safety systems, Johannes Heidecke, has informed staff that he is leaving the company. His departure comes amid a broader effort to integrate the safety and research functions within the organization.
The news was first reported by WIRED after viewing an internal memo from chief research officer Mark Chen. In the note, Chen explained that safety teams will now report to Mia Glaese, who holds the dual role of vice president of research and head of alignment, and will take on an expanded title as vice president of research and safety.
Details
The reorganization aims to align safety work more closely with model development cycles. As part of the change, Saachi Jain, who previously led safety teams at OpenAI, will serve as the interim head of safety systems, reporting directly to Glaese.
- Johannes Heidecke is stepping down as head of safety systems.
- Safety teams will now report to Mia Glaese, VP of research and head of alignment.
- Saachi Jain will act as interim head of safety systems.
- The shift follows a memo from chief research officer Mark Chen.
Quotes
In the memo, Mark Chen said, “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.” This statement underscores the pressure on the safety function as OpenAI accelerates its model releases.
Background
OpenAI has been under increasing scrutiny to ensure its AI systems are safe and aligned with human intentions. Over the past year, the company has experimented with different structures for its safety organization, including separate teams and joint reporting lines. The latest move reflects an attempt to streamline oversight while keeping safety considerations embedded in rapid development cycles.
Conclusion
Heidecke’s exit marks a notable change in OpenAI’s safety leadership at a time when the firm is pushing out new models more frequently. How the interim arrangement under Saachi Jain and the revised reporting to Mia Glaese will affect safety outcomes remains to be seen, but the reorganization signals OpenAI’s intent to keep safety tightly coupled with its research agenda.
Discover more from NewzQuest
Subscribe to get the latest posts sent to your email.
