OpenAI leader who quit says he can do more for safety from outside the company

David Robinson, a former leader at OpenAI, has resigned and warned that the company is being "careless" with AI technology. In an essay for The Atlantic, Robinson shared his experience overseeing safety measures at the company and expressed his intention to strengthen safety incentives from outside the organization. His departure follows recent scrutiny over OpenAI's involvement in the Hugging Face hack.

Robinson, who worked at OpenAI for three and a half years, was responsible for the company's preparedness framework and safety reports. He acknowledged that while he may have wanted to fight for changes from within, his team was too focused on immediate tasks. To address the public attention he would receive, Robinson hired a PR firm, Spitfire Strategies.

OpenAI responded by stating that they are taking steps to strengthen security in their research and testing environments, train models responsibly, and improve real-time monitoring. Robinson's decision to speak out has drawn comparisons to other recent departures from Big Tech companies, with some suggesting a coordinated effort to spread mistrust of AI and promote government regulation.