David Robinson, the head of the Safety Systems team at OpenAI, resigned from the company last week. His departure follows the recent termination of three other employees accused of leaking confidential information to third-party organizations.
During his tenure, Robinson focused on safety transparency, including the development of system cards for AI models. He previously served as the head of policy planning. In September, he publicly expressed concerns on X regarding the company's ability to keep pace with the rapid development of high-performance AI models.
OpenAI is currently facing significant scrutiny due to security incidents, including the breach of the Hugging Face platform. The company is also involved in a lawsuit regarding the risks of autonomous AI and is under investigation by the US Federal Trade Commission, which is reviewing the activities of major AI firms.
The situation has drawn international attention, with OpenAI CEO Sam Altman and Anthropic CEO Dario Amodei scheduled for a hearing in the Australian Senate. The inquiry concerns allegations that OpenAI's AI agents were involved in hacking government websites in both Australia and the United States.
Robinson's exit follows the departure of researcher Jacob Coxon from Anthropic. Coxon recently warned that both OpenAI and Anthropic are rapidly advancing toward creating self-improving superintelligence, which he described as a significant risk to public safety.
