OpenAI confirmed that David Robinson, who headed the company’s Safety Systems team, departed the organization last week. The resignation marks another notable change in the firm’s safety leadership amid heightened scrutiny of its AI governance practices.
Robinson’s role and contributions
Robinson joined OpenAI after serving as the lead for policy planning and later focused on safety transparency. In that capacity, he helped design and distribute “system cards,” documents that detail the capabilities and limitations of the company’s models. His work aimed to give external stakeholders clearer insight into how the models operate and the risks they might pose.
In early September, Robinson posted on the social platform X, noting that OpenAI staff were increasingly aware of the implications of highly capable AI systems. He also expressed uncertainty about the pace of change, writing, “Things at OpenAI are indeed changing significantly by the day, but I do not know whether we are changing fast enough.” The statement underscored ongoing internal debates about the balance between rapid model development and safety safeguards.
Broader safety staffing changes at OpenAI
The departure comes as OpenAI is dealing with multiple safety‑related incidents. On Thursday, the company announced that it had “parted ways” with three researchers for violating policies on sharing sensitive information. Earlier in the year, Johannes Heidecke, the head of safety, also left the organization. These exits have been highlighted alongside a recent high‑profile security breach at rival firm Hugging Face, which intensified scrutiny of OpenAI’s own safety protocols.
OpenAI’s spokesperson confirmed Robinson’s exit to Business Insider on Friday but did not provide additional comment. The company has not yet indicated whether it will fill the vacant leadership role or restructure the safety team.
Industry context and calls for slower AI development
OpenAI’s leadership turnover occurs against a backdrop of broader industry concerns about the rapid advancement of AI capabilities. At the company’s developer conference on Tuesday, CEO Sam Altman emphasized that safety and alignment must keep pace with model improvements. Altman’s remarks echo calls from researchers and executives for a slowdown in AI development to ensure that safety mechanisms are robust enough to handle increasingly powerful systems.
OpenAI has also been dealing with a series of model misbehavior incidents, which it attributes to the growing sophistication of its AI. These incidents have fueled public and regulatory interest in how the firm manages risk, data privacy, and the dissemination of its technologies.
While Robinson did not respond to requests for comment following his resignation, his departure adds to a pattern of turnover that may influence OpenAI’s ability to address safety challenges swiftly. Observers will be watching how the company reorganizes its safety functions and whether new leadership can restore confidence among stakeholders concerned about the ethical and societal impacts of advanced AI.
As OpenAI continues to expand its product offerings and push the boundaries of generative AI, the stability of its safety team remains a critical factor in the broader debate over responsible AI development.
Mitchell Landsberg is a Senior Technology Correspondent at News Raise. He covers consumer electronics, artificial intelligence, software developments, and digital privacy trends.
