Helen Toner
Last reviewed
Jun 9, 2026
Sources
15 citations
Review status
Source-backed
Revision
v2 · 2,352 words
Improve this article
Add missing citations, update stale details, or suggest a clearer explanation.
Suggest edit
CtrlK
Last reviewed
Jun 9, 2026
Sources
15 citations
Review status
Source-backed
Revision
v2 · 2,352 words
Add missing citations, update stale details, or suggest a clearer explanation.
Helen Toner is an Australian researcher who works on the policy and governance of artificial intelligence. She is the interim executive director of the Center for Security and Emerging Technology (CSET) at Georgetown University, where she previously served as director of strategy and foundational research grants. [1][2] Toner is widely known beyond policy circles for her time on the nonprofit board of OpenAI, which she joined in 2021 and left in late 2023 after the board briefly removed and then reinstated chief executive Sam Altman. [3][4] Since that episode she has written and spoken publicly about AI governance, corporate accountability, and the limits of industry self-regulation. [5][6]
Much of Toner's work centers on how governments and institutions can oversee a technology whose capabilities are hard to measure and hard to predict. She has testified before the United States Congress, contributed to outlets such as Foreign Affairs and the Economist, and delivered a talk on the subject at the TED conference in 2024. [1][7][8] In 2024 she was named to TIME magazine's list of the 100 most influential people in AI. [9]
Toner was born in 1992 and grew up in Melbourne, Australia. [3] She was introduced to the effective altruism movement as a student, an exposure that shaped her later interest in the long-run social effects of powerful technologies. [3] She studied at the University of Melbourne, completing a Bachelor of Science in chemical engineering along with a diploma in languages. [1][3] She later earned a Master of Arts in security studies from Georgetown University's Walsh School of Foreign Service. [1]
Early in her career Toner worked in the effective altruism and grantmaking world, including at the charity evaluator GiveWell and then at Open Philanthropy, where she was a senior research analyst advising policymakers and grantmakers on AI policy and strategy. [1][3] Open Philanthropy was the principal funder behind CSET, providing a multiyear founding grant to establish the center at Georgetown. [3]
In 2018 Toner spent roughly nine months living in Beijing, where she studied Mandarin and researched China's AI ecosystem as a research affiliate of the University of Oxford's Centre for the Governance of AI. [1][3] That period informed a recurring strand of her later analysis on how China develops and deploys AI and on what the comparison means for technology competition and national security. She has co-written on the subject for Foreign Affairs. [1][3]
CSET is a policy research organization, or think tank, housed within Georgetown's Walsh School of Foreign Service. It studies the security implications of emerging technologies, with a particular focus on artificial intelligence, and produces analysis intended to inform government decision-makers. [1] Toner joined CSET in January 2019, near its founding, as director of strategy. [3] In March 2022 she became its full-time director of strategy and foundational research grants, a role in which she helped set the center's long-run priorities and oversaw a multimillion-dollar program of technical research grants. [2][3]
In August 2025 CSET announced that Toner would become its interim executive director, effective in early September 2025, succeeding Dewey Murdick. [2] Joel Hellman, dean of the School of Foreign Service, said in the announcement that Toner had been instrumental in building the center, and Toner said she was honored to step into the role at a notable moment for AI and other emerging technologies. [2]
Toner's research and commentary span several recurring themes. She argues that AI capabilities are difficult to measure and that better methods to evaluate what systems can do are a precondition for sound policy. [7] She has favored disclosure and transparency requirements for developers of advanced systems, including information about what models can do, what data they are trained on, and how companies manage risks. [7] She has also supported external auditing of AI systems by independent parties rather than leaving safety assessment to the companies that build the models, a position she has summarized with the phrase that firms should not be left to grade their own homework. [7] Her writing engages with the prospect of artificial general intelligence and with AI safety concerns, while emphasizing that policymakers may need to act even without scientific consensus about how the systems work or what will be built next. [8]
Toner joined the board of directors of OpenAI in 2021. [3][4] OpenAI is governed by a nonprofit whose board holds legal authority over the organization, and Toner was one of several independent directors. According to multiple accounts she was invited to the board by Holden Karnofsky of Open Philanthropy, who left the seat as she joined. [3] During her tenure the board included Altman, co-founder and chief scientist Ilya Sutskever, Quora chief executive Adam D'Angelo, and entrepreneur Tasha McCauley, among others. [4]
On November 17, 2023, OpenAI's board announced that it had removed Altman as chief executive. [4] The board's public statement said Altman had not been consistently candid in his communications with the board, which it said hindered its ability to exercise its responsibilities, and that the board no longer had confidence in his continued leadership. [4] Four directors voted for the removal: Toner, Sutskever, McCauley, and D'Angelo. [4] President and chairman Greg Brockman, who had not been told in advance, stepped down from his role and left the company along with several senior researchers. [4]
The decision set off several days of turmoil. Microsoft, OpenAI's largest investor and partner, said it would hire Altman and Brockman to lead a new AI group, and a letter signed by a large majority of OpenAI's roughly 770 employees, reported at the time as more than 700 staff, threatened mass resignation unless the board reversed course and resigned. [4][10] Sutskever, who had supported the removal, publicly said he regretted his role. [4] Mira Murati and then Emmett Shear served as interim chief executives during the interval. [4] On November 22, 2023, OpenAI announced that Altman would return as chief executive under a new initial board chaired by Bret Taylor and including Larry Summers and the continuing director Adam D'Angelo. [10] As part of that reconstitution Toner and McCauley left the board. [10] Toner said she formally resigned on November 29, 2023. [11]
The board did not give a detailed public account of its reasoning at the time, and competing explanations circulated. Some reporting linked the conflict in part to a research paper Toner had co-authored at CSET in October 2023, titled "Decoding Intentions," which discussed how AI developers signal their intentions and compared OpenAI's public messaging unfavorably with that of rival Anthropic in places. [3][12] According to later accounts, Altman objected to the paper, and Toner has said that the disagreement fed into a broader loss of trust. [3][13] OpenAI and Altman's supporters, for their part, characterized the removal as poorly explained and unjustified, and the overwhelming employee response reflected support for Altman's return. [4][10]
The reconstituted board commissioned an independent review of the events from the law firm WilmerHale. OpenAI announced the review's completion on March 8, 2024. [14] According to OpenAI's statement, the firm conducted dozens of interviews and reviewed more than 30,000 documents, and it concluded that the prior board's decision did not arise out of concerns about product safety or security, the pace of development, OpenAI's finances, or its statements to investors, customers, or business partners, but rather followed from a breakdown in the relationship and a loss of trust between the prior board and Altman. [14] OpenAI said the review found that Altman's conduct did not mandate removal and that the board unanimously concluded he and Brockman were the right leaders for the company. [14] Altman returned to the OpenAI board, and three new directors were added. [14]
Toner spoke at greater length about the episode in 2024, after the review closed. In a May 2024 interview on the TED AI Show podcast she said the board had concluded it could not trust information that Altman provided. [5][13] She said the board had not been told in advance about the launch of ChatGPT in November 2022 and learned of it from social media, that Altman had not disclosed his ownership of the OpenAI Startup Fund, and that he had given the board inaccurate information about the company's formal safety processes. [5][13] She also said two company executives had described to the board what she characterized as a toxic culture and behavior they likened to psychological abuse, supported by documentation. [5][13] Toner said the dispute was not fundamentally about AI safety but about a pattern she described as withholding information, misrepresenting events, and in some cases lying to the board, which she said made the board's oversight role unworkable. [5][13]
In an op-ed in the Economist published in May 2024, Toner and McCauley wrote that they stood by the decision and argued that their experience had convinced them that self-governance could not reliably withstand the pressure of profit incentives, and that governments needed to establish effective regulatory frameworks for AI. [6] They also cautioned that poorly designed rules could entrench incumbents and burden smaller firms. [6]
OpenAI and members of its new board disputed the former directors' characterization. Board chair Bret Taylor and Larry Summers said they did not accept the claims made by Toner and McCauley about events at OpenAI, and Taylor said the company was disappointed that Toner continued to revisit the matter. [6][13] OpenAI noted the WilmerHale review's findings, including that the removal had not been driven by safety, financial, or product concerns. [14] Observers across the press and the AI field reached differing conclusions about the relative weight of the various accounts, and the episode remained contested. [4][13]
In the period after she left the board, Toner became a frequently cited voice on AI policy. Her TED talk in 2024, titled "How to govern AI even if it's hard to predict," argued that uncertainty about AI should not be a reason to avoid governance, and set out proposals around measuring capabilities, requiring disclosure from developers, and using independent audits. [7] She has tied these arguments to her board experience, contending that internal guardrails at a company can be fragile when large amounts of money are at stake and that external oversight is therefore needed. [8]
Toner has testified before Congress on several occasions. In written testimony to the Senate Judiciary Committee in 2024, she urged lawmakers to set transparency requirements, to clarify how liability for AI harms should be allocated so that companies face an economic incentive toward a duty of care, and to prepare for the possibility of highly capable systems. [8] She wrote that her time on OpenAI's board had taught her how fragile internal guardrails can be when money is on the line, which she said was a reason for policymakers to act. [8] Her broader writing situates AI governance within questions of national security and technology competition, drawing on her earlier research at CSET on China and on export controls. [1]
Toner appeared on TIME magazine's inaugural and subsequent lists of the 100 most influential people in AI, including the 2024 edition, which highlighted her role in the OpenAI board episode and her advocacy for AI oversight. [9] Profiles in outlets such as Fast Company described her as becoming a more prominent advocate for responsible AI after leaving OpenAI rather than a diminished figure. [15] Through CSET, her congressional testimony, her commentary in Foreign Affairs and the Economist, and her public talks, she has remained an active participant in debates over how AI should be regulated and how AI companies should be held accountable. [1][6][8]
| Field | Detail |
|---|---|
| Full name | Helen Toner |
| Born | 1992, Melbourne, Australia |
| Nationality | Australian |
| Education | B.Sc. chemical engineering and diploma in languages, University of Melbourne; M.A. security studies, Georgetown University |
| Known for | AI governance research; OpenAI board membership and the November 2023 removal of Sam Altman |
| Current role | Interim executive director, Center for Security and Emerging Technology (CSET), Georgetown University |
| Prior CSET role | Director of strategy and foundational research grants |
| Earlier roles | Senior research analyst, Open Philanthropy; research affiliate, Oxford Centre for the Governance of AI (Beijing) |
| OpenAI board | Joined 2021; resigned November 29, 2023 |
| Recognition | TIME 100 most influential people in AI (2024) |