Every AI Policy Ever · open in the interactive register

Published (non-binding, current)Non-binding · Institutional mandate

Tackling AI security risks to unleash growth and deliver Plan for Change (renaming of the AI Safety Institute to the AI Security Institute)

Record GB-07

Issued by
Department for Science, Innovation and Technology (DSIT)
Jurisdiction
United Kingdom
Level · region
National · Europe
Published / adopted
2025-02-14
Language
English

Summary

Renames the AI Safety Institute the AI Security Institute and focuses it on serious AI risks with security implications, such as the use of AI to develop chemical and biological weapons, to carry out cyber-attacks and to enable crimes such as fraud and child sexual abuse. The institute will not focus on bias or freedom of speech. It gains a criminal misuse team working jointly with the Home Office, and works with the Defence Science and Technology Laboratory, the Laboratory for AI Security Research, the National Cyber Security Centre and the national security community to assess the risks of frontier AI.

Safety topics

safety institutenational security & militaryevaluation & testingchild safetycatastrophic riskgeneral-purpose / frontier AI

Status notes

Announced by Technology Secretary Peter Kyle at the Munich Security Conference on 14 February 2025, days after the AI Action Summit in Paris. The institute's website gives its mission as 'to equip governments with a scientific understanding of the risks posed by advanced AI' and describes it as a research organisation within DSIT.

Sources

Open the official source https://www.gov.uk/government/news/tackling-ai-security-risks-to-unleash-growth-and-deliver-plan-for-change

Secondary source: https://www.aisi.gov.uk/

Local copy

Archived. Last attempt 2026-10-18 06:48:43 UTC

Official source (third-party copy, not published) archive/GB-07/2026-10-11/official.html
Retrieved 2026-10-11 14:03:26 UTC
SHA-256 c9aecd756c124a9be46124e850183d0b925cdd6620b86051f9a8b30a2ef746eb
Secondary source (third-party copy, not published) archive/GB-07/2026-10-11/secondary.html
Retrieved 2026-10-11 14:03:26 UTC
SHA-256 d1ff6a24dc9fa34e4d9cd4178c5cfa617310c5a070476f6f3cb31db3150a1805

Verification

B
Another official source, such as an announcement by the issuer, confirms the title, date and status. checked 2026-10-18 · alternative-official-url-fetched
Refresh of 18 Oct 2026 under the rule that level A needs the instrument's own text or an official register. Re-read the GOV.UK press release of DSIT, the AI Security Institute and Peter Kyle, published 14 February 2025 (stored 11 Oct 2026; the archiver found it unchanged on 18 Oct 2026), and the AI Security Institute's home page (aisi.gov.uk, stored as the secondary source; mission and place within DSIT). The press release confirms the title, the date, the renaming and the institute's new focus; the home page shows the institute operating under the new name. The record rests on the press release, which is an announcement of the decision: no separate text of the decision to rename and refocus the institute is among the stored files, and the press release links none. Its quotes therefore count as alt-official, not as the instrument's own text. GOV.UK marks the release as published under the 2024 to 2026 Starmer Labour government.
Second check: Independent check by Claude Code (claude-sonnet-5-5), 2026-10-18: 5 field(s) supported.

History

  1. 2026-10-18related: [{"slug": "uk-ai-safety-institute-2023", "id": "GB-03"}] → [{"slug": "uk-ai-safety-institute-2023", "id": "GB-03"}, {"id": "GB-05", "slug": — two-way link added: GB-05, GB-09, INT-19 link(s) to GB-072026-10-18-fix-related-links · Claude Code (claude-opus-5-5)
  2. 2026-10-18verification.level: A → B — level computed from the evidence by tools/pipeline.py2026-10-18-refresh-level-rule-1 · Claude Code (claude-opus-5-5)
  3. 2026-10-17notes: Announced by Technology Secretary Peter Kyle at the Munich Security Conference o → Announced by Technology Secretary Peter Kyle at the Munich Security Conference o — 'The institute is not a regulator' removed: it is in GB-03's source, not in this record's stored files; the AI Action Summit added2026-10-17-recheck-europe-2 · Claude Code (claude-opus-5-5)
  4. 2026-10-17summary: Renames the AI Safety Institute as the AI Security Institute and refocuses it on → Renames the AI Safety Institute the AI Security Institute and focuses it on seri — worded as the press release words it: 'uplift' and 'explicitly out of scope' replaced by the release's own terms; the Laboratory for AI Security Research added2026-10-17-recheck-europe-2 · Claude Code (claude-opus-5-5)

Related records

UK AI Safety Institute founding document United Kingdom · Superseded

UK AI Opportunities Action Plan United Kingdom · Published (non-binding, current)

Crime and Policing Act 2026 (AI provisions) United Kingdom · Partially in force

Five Country Ministerial Communiqué 2026 (AI and national security) Five Country Ministerial (Australia, Canada, New Zealand, United Kingdom, United States) · Published (non-binding, current)

Citation

Department for Science, Innovation and Technology (DSIT). Tackling AI security risks to unleash growth and deliver Plan for Change (renaming of the AI Safety Institute to the AI Security Institute). 2025-02-14. https://www.gov.uk/government/news/tackling-ai-security-risks-to-unleash-growth-and-deliver-plan-for-change. Status: Published (non-binding, current); verified 2026-10-18 (level B). Local copy archive/GB-07/2026-10-11/official.html retrieved 2026-10-11 14:03:26 UTC, SHA-256 c9aecd756c124a9be46124e850183d0b925cdd6620b86051f9a8b30a2ef746eb. Every AI Policy Ever, version 1.18 (snapshot 2026-10-18), record GB-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.