What happened
On 18 November 2026 the House of Commons Business, Innovation, Science and Trade Committee published its Eleventh Report of Session 2026-27 (HC 689), 'Guardrails for growth: AI safety and security'. It identifies ten gaps in UK frontier-AI governance and recommends an AI Safety and Security Bill in the next parliamentary session plus a new statutory regulator - distinct from the AI Security Institute (AISI) - with legal powers to require independent verification of developers' safety claims, pause high-risk training and testing, delay or prohibit deployment, and suspend services where risks cannot be controlled. It also recommends mandatory incident/near-miss reporting, protected whistleblower channels, a clear liability regime, criminal offences for harmful misuse, and an international convention, with G20 2027 objectives within three months.
Why it matters
This is the first formal UK parliamentary push to convert AISI's voluntary testing into enforceable pre-release safety obligations. If implemented it would impose statutory model-access, verification, incident-reporting and stop/block powers on frontier and designated high-risk AI developers operating in the UK, materially changing release timelines and compliance posture.
Action needed
Frontier AI developers and large deployers in the UK should track the legislative roadmap and prepare to substantiate safety claims with evidence, support independent evaluator access, and build incident-reporting/whistleblower channels ahead of a likely AI Safety and Security Bill.