Helen Toner

Helen Toner (born 1992 in Melbourne) is an Australian AI-policy researcher at Georgetown’s Center for Security and Emerging Technology (CSET), where she has served as director of strategy and foundational research grants. She came up through the effective-altruism world of Open Philanthropy, studied China’s AI ecosystem, and in 2021 joined the nonprofit board of OpenAI — the body charged with keeping the company’s commercial engine subordinate to its safety mission.

The board member who said no

In November 2023 Toner and fellow directors voted to fire CEO Sam Altman, citing a loss of trust; after his return days later, she resigned. She later said publicly that Altman had given the board inaccurate information for years — including that it learned about ChatGPT’s launch from Twitter — and the episode turned her into the leading witness for the claim that self-governance at frontier labs fails under commercial pressure.

Testifying before the Senate Judiciary subcommittee in September 2024, she noted that top companies including OpenAI, Google, and Anthropic are building AGI “as an entirely serious goal,” warned that the science of measuring AI risks is extremely immature, and argued for transparency requirements, whistleblower protections, and third-party audits. Skeptics of her account note the fired CEO was reinstated with employee support and that the board never fully detailed its reasoning; Toner has answered with steadily more specific public disclosures, in essays, testimony, and her podcast on AI policy.