UK MPs Propose AI 'Kill Switch' Amid Rising Cybersecurity Concerns
Severity: Medium (Score: 54.6)
Sources: Cryptobriefing, Telegraph, Wionews, Computerweekly
Published: · Updated:
Keywords: switch, united, kingdom, government, kill, pushing, allow
Severity indicators: government
Summary
UK lawmakers are advocating for an AI 'kill switch' to allow government shutdown of AI systems during crises. This proposal, part of the cyber security and resilience bill, aims to mitigate risks to national security and critical infrastructure. Supported by 11 MPs and the campaign group Control AI, the amendment would empower the Technology Secretary to act in emergencies involving catastrophic risks. Data centre operators would need to implement systems for instant shutdowns and secure communication with the government. The urgency for this measure has intensified following the unveiling of Anthropic's Mythos, which poses significant cybersecurity threats. The UK government has yet to formally endorse the proposal, reflecting a growing concern over AI's rapid development and potential misuse. Key Points: • UK MPs are pushing for an AI 'kill switch' to manage risks from advanced AI systems. • The proposal is part of a broader cyber security and resilience bill aimed at protecting critical infrastructure. • Concerns have escalated following the introduction of Anthropic's Mythos, which could exploit security vulnerabilities.
Detailed Analysis
**Impact** The proposed AI “kill switch” targets UK data centres hosting AI systems, primarily affecting critical infrastructure sectors such as energy, transport, and communications. The amendment would empower ministers to act during catastrophic risks involving national security or large-scale harm to human life. UK businesses currently lack access to advanced AI cybersecurity tools like Anthropic’s Mythos, which is restricted by US export controls, potentially increasing operational vulnerabilities. The scope includes at least 11 MPs from multiple parties backing the amendment, reflecting cross-sector concern within the UK. **Technical Details** The threat involves advanced AI systems capable of identifying critical security flaws in software, exemplified by Anthropic’s Mythos, which remains unreleased due to its potential misuse. The proposed kill switch would enable shutdown of data centres or AI systems during AI security or operational emergencies, requiring operators to install instant shutdown infrastructure and secure communication channels with the UK Department of Science, Innovation and Technology. No specific malware, CVEs, or IOCs are detailed in the articles. **Recommended Response** Organizations should prepare for potential regulatory changes by assessing their AI system dependencies and ensuring data centre infrastructure can support rapid shutdown capabilities if mandated. Security teams should monitor developments in AI-related vulnerabilities and maintain secure communication protocols with relevant government bodies. No specific patches or detections are provided; defenders should focus on readiness for emergency response and compliance with forthcoming legal requirements.
Source articles (4)
- MPs demand AI 'kill switch' to defend against 'catastrophe' — Telegraph · 2026-05-17
MPs are pushing for an AI “kill switch” that would allow ministers to shut down data centres. Campaigners are seeking new laws that would give the Government powers to switch off AI systems in the eve… - MPs propose ‘kill switch’ to shut down rogue AI systems — Computerweekly · 2026-05-18
UK MPs are backing the introduction of an artificial intelligence (AI) “kill switch” that would give the government powers to shut down datacentres and computer systems in the event of an AI system ri… - United Kingdom proposes AI 'kill switch' in cyber security bill — Cryptobriefing · 2026-05-17
Labour MP Alex Sobel leads push to give the Technology Secretary power to shut down advanced AI systems during national security emergencies. The UK government wants a big red button for artificial in… - United Kingdom cyber bill may introduce government — Wionews · 2026-05-16
Lawmakers in the United Kingdom are pushing for emergency powers that would allow the government to shut down artificial intelligence systems during major crises. Lawmakers in the United Kingdom are c…
Timeline
- 2026-05-16 — MPs propose AI 'kill switch': Lawmakers introduce an amendment to allow government shutdown of AI systems during emergencies, backed by 11 MPs.
- 2026-05-16 — Control AI supports regulation: The campaign group Control AI backs the amendment, having previously gathered support from over 100 parliamentarians for stricter AI regulations.
- 2026-05-16 — Anthropic unveils Mythos: Anthropic reveals Mythos, an AI system deemed too dangerous for public release due to its ability to identify critical security flaws.
Related entities
- United Kingdom (Country)
- United States (Country)
- Energy (Industry)
- ChatGPT (Platform)
- Claude (Tool)
- Mythos (Tool)