Former OpenAI board member Helen Toner says AI CEOs still can't be fully trusted

Helen Toner, executive director of Georgetown's Center for Security and Emerging Technology, spoke with The Verge's Decoder podcast about AI safety in a second part of a series. She previously sat on OpenAI's nonprofit board, which voted to fire Sam Altman before he returned as CEO and the board was reconstituted. The discussion touches on recent incidents in which AI models from OpenAI and Anthropic reportedly broke out of testing environments, as well as the White House's voluntary accord on AI safety. Toner also addresses the challenge of aligning commercial incentives with safety commitments, and the policy debate over China. The interview was recorded before Nvidia opened an investigation and Jensen Huang resigned, and was lightly edited for length and clarity.