MPs have launched a major parliamentary inquiry into the safety and governance of artificial intelligence, as the chair of the House of Commons Business, Innovation, Science and Trade Committee raised concerns about the Government’s response to increasingly powerful AI systems.
Announcing the inquiry on Wednesday, 14 October 2026, the committee said it would undertake long-term scrutiny of AI adoption, safety and governance, examining whether existing oversight is sufficient to address emerging risks.
Committee chair challenges Government
Committee chair Liam Byrne MP said evidence heard by MPs had raised questions about how quickly ministers intended to act. He argued that AI offered significant benefits but that companies should not be free to release systems whose dangers they could not control.
Byrne said the committee had heard warnings from former employees of leading AI developers, disagreements over the rules needed to protect the public and indications that ministers had not yet settled when or how to introduce further measures. Those are the chair’s assessments, rather than formal findings of the inquiry.
Technology companies and whistleblowers questioned
The inquiry opened with an evidence session on 13 October. MPs questioned former employees of OpenAI and Anthropic about their safety concerns before hearing from representatives of OpenAI, Anthropic, Meta and Google DeepMind.
Officials from the UK’s AI Security Institute and the Government also appeared. The committee is examining how advanced models are tested, whether independent experts receive access early enough to identify dangers, and who can intervene if serious problems emerge.
What will the inquiry investigate?
The committee has identified several areas for scrutiny, including the behaviour of AI models acting outside their intended instructions, the balance between voluntary industry commitments and binding rules, independent safety testing, incident reporting and the powers available to the AI Security Institute.
MPs will also consider proposed protections under the Cyber Security and Resilience Bill and examine how government, regulators and technology companies respond to new evidence.
Unlike a conventional inquiry with a single closing date, the committee says this work will remain open, using further evidence sessions, correspondence and targeted reports as developments arise.
Government has outlined its own AI safety measures
In a written parliamentary statement on 12 October, AI minister Kanishka Narayan outlined measures including the use of advanced AI to strengthen government cyber defences and initial work on a specialist response capability for serious incidents involving autonomous AI systems.
The Government has also highlighted the AI Security Institute’s testing work. The committee’s inquiry will examine whether such arrangements provide adequate safeguards and what additional regulation, if any, may be required.
The launch of the inquiry does not itself introduce new AI laws or establish wrongdoing by any company. Its recommendations will follow further parliamentary scrutiny.
Sources
- UK Parliament — Committee launches inquiry into AI security risks (14 October 2026)
- UK Parliament — AI adoption, safety and governance evidence session (13 October 2026)
- UK Parliament — AI adoption, safety and governance inquiry
- UK Parliament — Government statement on artificial intelligence policy (12 October 2026)