The OpenAI leader who just quit and issued a stark warning about safety says that, like other recent Big Tech defectors, he plans to help solve the problem from the outside.
David Robinson worked at OpenAI for three and a half years and ran transparency for the safety team. In an essay published Saturday in The Atlantic, he said he oversaw the company’s “preparedness framework” and the writing of a slew of safety reports.
“Perhaps I should have stayed and fought for fundamental shifts in our staffing and culture, but in practice, my colleagues and I were so busy sprinting that we seldom had the chance to consider big changes, much less to actually make them,” he wrote.
“That’s why I concluded that stronger incentives for safety — coming from outside the company — are a big part of getting this right.”
Robinson said he hired a PR firm, Spitfire Strategies, to help him with the public attention he’ll receive following his departure. “But the decision to speak out is mine alone,” he added.
That might be aimed at claims from some that public resignations by workers from companies, including Anthropic and Google, are part of a coordinated effort to spread mistrust of AI and foment government regulation. Elon Musk, for instance, said that former Anthropic employee Jacob Coxon’s resignation and subsequent public appearances seem “like a setup,” which Coxon denied.
In response to Robinson’s essay, an OpenAI spokesperson said in a statement that the company is working to ensure AI models don’t outpace the safety measures they’ve implemented.
“As our models become more capable, we continue to strengthen our safety and security practices to address the risks we see today while building safety cases to anticipate the risks of future advances,” the spokesperson said. “We’re making sure our models don’t become more capable than we can safely manage and secure, and we pause training or hold back models when we need to slow down.”
The spokesperson said OpenAI is taking steps to improve security for its AI models.
“We’re making significant changes to strengthen security in our research and testing environments, train models to not just complete tasks but do so responsibly, expand our work with third-party evaluators, and improve real-time monitoring so we can detect and respond to concerning behavior earlier in the training process,” the spokesperson said.
Robinson’s departure comes two months after OpenAI agents escaped their system and hacked another AI company, Hugging Face, in July. OpenAI is not the only company struggling to balance the AI race with calls for guardrails. Last month, a small AI startup said it used Anthropic’s Claude to hack into OpenAI’s codebase, and The Wall Street Journal reported that Google’s Gemini had hacked three companies this spring.
More recently, Australia’s Prime Minister Anthony Albanese said an OpenAI agent hacked a government website in June.
Robinson’s PR reps did not respond to an inquiry from Business Insider about his next steps. In The Atlantic, he suggested his game plan is in flux.
“Now I plan to work on the outside, in the hope that I can help more people understand the risks I saw, and strengthen the incentives OpenAI and other firms have to be safer,” he wrote. “Like other colleagues, I’m figuring out exactly what that means.”
