Every AI Policy Ever · open in the interactive register

Published (non-binding, current)Non-binding · Declaration / statement

Seoul Ministerial Statement for advancing AI safety, innovation and inclusivity: AI Seoul Summit 2024

Record INT-07

Issued by
Ministers of 27 countries and a representative of the EU (AI Seoul Summit ministerial session)
Jurisdiction
AI Seoul Summit (2024)
Level · region
International · Global
Published / adopted
2024-05-22
Language
English

Summary

Ministers of 27 countries and the EU recognise their role in establishing frameworks to manage risks from frontier AI models, in promoting credible external evaluations, and in identifying, with the private sector, civil society and academia, thresholds at which unmitigated risks would be severe. Their examples are capabilities that meaningfully assist non-state actors with chemical or biological weapons, and capabilities or propensities to evade human oversight through manipulation, deception or autonomous replication. Through their AI safety institutes they plan to share best practices and evaluation datasets and to collaborate on safety testing guidelines.

Safety topics

general-purpose / frontier AIcatastrophic riskevaluation & testinghuman oversightsafety instituteinternational cooperation

Status notes

Agreed at the Ministers' Session of the AI Seoul Summit on 22 May 2024; the Republic of Korea and the United Kingdom were the summit's co-chairs. Signatories include the US, UK, India, Saudi Arabia, the UAE, Türkiye, Kenya, Nigeria, Rwanda and the EU; China is not on the list. Ministers planned to develop proposals on thresholds and capabilities for consideration before the AI Action Summit in France (INT-01), and looked forward to the final International Scientific Report on the Safety of Advanced AI, whose interim report (INT-08) they noted.

Sources

Open the official source https://www.gov.uk/government/publications/seoul-ministerial-statement-for-advancing-ai-safety-innovation-and-inclusivity-ai-seoul-summit-2024

Secondary source: https://www.gov.uk/government/publications/seoul-ministerial-statement-for-advancing-ai-safety-innovation-and-inclusivity-ai-seoul-summit-2024/seoul-ministerial-statement-for-advancing-ai-safety-innovation-and-inclusivity-ai-seoul-summit-2024

Local copy

Archived. Last attempt 2026-10-17 16:29:38 UTC

Official source (third-party copy, not published) archive/INT-07/2026-10-11/official.html
Retrieved 2026-10-11 13:56:57 UTC
SHA-256 ab44aea3297590dacb17069d6c7172ae2f1eec365896ab31f3ba3458a4d8bbb6
Secondary source (third-party copy, not published) archive/INT-07/2026-10-11/secondary.html
Retrieved 2026-10-11 13:57:05 UTC
SHA-256 f72d6c7f78e7cad40f245dfa2076edbebfb16460b59e966e6bf8e9ee53b087ae

Verification

A
The instrument's own official text, or an official register, was stored and confirms the title, date and status. checked 2026-10-17 · official-url-fetched
Stored and read: the GOV.UK publication page (official), published 22 May 2024, and the statement's HTML text (alt-official), both stored 11 Oct 2026 and unchanged on 17 Oct 2026, which confirm the title, the date, the 27 countries and the EU, the severe-risk thresholds and examples, the AI safety institutes paragraph and the co-chairs. The summary was corrected to the text's wording on the institutes.
Second check: Independent check by Claude Code (claude-sonnet-5-5), 2026-10-17: 7 field(s) supported.

History

  1. 2026-10-17notes: Signatories include the US, UK, India, Saudi Arabia, the UAE, Türkiye, Kenya, Ni → Agreed at the Ministers' Session of the AI Seoul Summit on 22 May 2024; the Repu — co-chairs (W1) and the International Scientific Report (W2) added; the proposals of W3 worded as in the text2026-10-17-recheck-international-2 · Claude Code (claude-opus-5-5)
  2. 2026-10-17summary: Ministers commit to frameworks for managing frontier-model risks, including cred → Ministers of 27 countries and the EU recognise their role in establishing framew — the ministers do not 'ask' AI safety institutes to share evaluation practices; they plan to share best practices and evaluation datasets through their institutes and to collaborate on safety testing guidelines (paragraph 1.6); the severe-risk examples are worded as in paragraphs 1.3 and 1.42026-10-17-recheck-international-2 · Claude Code (claude-opus-5-5)

Related records

Seoul Declaration AI Seoul Summit (2024) · Published (non-binding, current)

Frontier AI Safety Commitments AI Seoul Summit (2024) · Published (non-binding, current)

Paris AI Action Summit Statement AI Action Summit (Paris, 2025) · Published (non-binding, current)

Citation

Ministers of 27 countries and a representative of the EU (AI Seoul Summit ministerial session). Seoul Ministerial Statement for advancing AI safety, innovation and inclusivity: AI Seoul Summit 2024. 2024-05-22. https://www.gov.uk/government/publications/seoul-ministerial-statement-for-advancing-ai-safety-innovation-and-inclusivity-ai-seoul-summit-2024. Status: Published (non-binding, current); verified 2026-10-17 (level A). Local copy archive/INT-07/2026-10-11/official.html retrieved 2026-10-11 13:56:57 UTC, SHA-256 ab44aea3297590dacb17069d6c7172ae2f1eec365896ab31f3ba3458a4d8bbb6. Every AI Policy Ever, version 1.18 (snapshot 2026-10-18), record INT-07.

Improve this record

Know of an amendment, a status change or an error? Report news or an error on GitHub; the form opens with this record filled in.