Rishi Sunak
AI-generated text. This page was generated using artificial intelligence.
Rishi Sunak served as United Kingdom prime minister from October 25, 2022 to July 5, 2024. His AI-policy work includes the 2023 Bletchley Park AI Safety Summit and the creation of the UK's AI Safety Institute, followed by advocacy for stronger oversight of frontier research.[1][2][3]
Relevant roles and affiliations
Sunak's government role placed him at the center of the UK's international AI-safety diplomacy. In October 2023 he announced plans for government model evaluation and defended inviting China to Bletchley, arguing that a serious international strategy required engagement with leading AI powers.[4]
In September 2026 he identified himself as a senior adviser at Anthropic. The 2025 business-appointment advice describes a paid, part-time, internally focused role advising on strategic, macroeconomic and geopolitical matters, with payments to be donated to The Richmond Project. The committee imposed restrictions on use of privileged information and specified two-year restrictions, measured from his departure from ministerial office, on lobbying, government engagement and certain advice. Its advice was not an endorsement of the appointment. These are the terms of the dated advice, not a claim that every time-limited condition remains in force.[5][3]
Contributions and positions
Government evaluation and Bletchley
In his October 26, 2023 speech, Sunak paired optimism about AI's benefits with concern about misuse and possible loss of human control. He acknowledged uncertainty and disagreement about extreme risks. He favored building the government's ability to understand and test models before rushing into regulation, and argued that developers should not be the sole judges of their systems' safety.[4]
The AI Safety Institute launched on November 2, 2023 as an evolution of the Frontier AI Taskforce, with Ian Hogarth continuing as chair. Its announced work included testing frontier models before and after release and collaboration with international partners. Sunak presented it as a hub for research on capabilities and risks. The launch established a research and evaluation institution; the announcement should not be treated as a statutory model-licensing regime.[2] For the institute's subsequent development and the distinction between testing and regulation, see United Kingdom policy on catastrophic AI risk.
September 2026: stronger frontier oversight
On September 12, 2026, Sunak supported Dario Amodei's call to slow frontier capability advances so safety work could keep pace. He argued that voluntary pre-release testing, appropriate in his view in 2023, was now insufficient. He called for governments, starting with the United States, to require independent verification of research safety, better information about laboratories' proximity to recursive self-improvement, and protection against model theft.[3]
Sunak distinguished faster adoption of existing AI capabilities from the pace of frontier research. He argued that the latter should not be left solely to laboratories. He also called for US–China dialogue about avoiding catastrophe while acknowledging a tradeoff with maintaining a technological advantage over China. His claims about accelerating self-improvement and reduced understanding of models are attributed risk assessments, not established forecasts or findings of this article. The post advocates policy changes; it does not announce an enacted law or an international agreement.[3]
Selected speeches and statements
- Speech on AI (October 26, 2023) — Official transcript of Sunak's argument for government testing and international cooperation.[4]
- Statement on frontier oversight (September 12, 2026) — Public policy advocacy supporting stronger verification and pacing.[3]
Institutional records and wiki coverage
- AI Safety Institute launch (November 2, 2023) — Government announcement, distinct from an independently authored assessment.[2]
- September 12, 2026: frontier pacing and oversight debate.[3]
Related articles
- People in AI and Law
- United Kingdom policy on catastrophic AI risk
- United Kingdom AI Law
- United States policy on catastrophic AI risk
- Dario Amodei
- Anthropic
References
- ↑ UK government, The Rt Hon Rishi Sunak MP, biography. Accessed September 12, 2026.
- ↑ 2.0 2.1 2.2 UK government, Prime Minister launches new AI Safety Institute, November 2, 2023. Accessed September 12, 2026.
- ↑ 3.0 3.1 3.2 3.3 3.4 3.5 Rishi Sunak, public post on frontier-AI oversight, X, September 12, 2026. Full public post reviewed September 12, 2026.
- ↑ 4.0 4.1 4.2 Rishi Sunak, Prime Minister's speech on AI, October 26, 2023, Royal Society. Official transcript; accessed September 12, 2026.
- ↑ Advisory Committee on Business Appointments, Advice Letter: Rishi Sunak, Senior Advisor, Anthropic PBC, updated October 9, 2025. Accessed September 12, 2026.