Towards the end of Keir Starmer's premiership, senior ministers began developing plans for a new AI safety law, including reviewing existing legislation and exploring mandatory safety testing for advanced technology products. These plans reportedly stalled amid political upheaval.
Upon taking office earlier this summer, Prime Minister Andy Burnham abolished the Department for Science, Innovation and Technology (DSit), which had been leading the legislative work. Kanishka Narayan, the junior minister for AI, was moved to the Cabinet Office without the dedicated departmental support.
This move has prompted concerns from some in the AI industry and parliament. Chi Onwurah, the Labour MP who chairs the science and technology committee, stated that the government's response to an existential threat cannot be to claim helplessness. Matt Clifford, an AI investor, posted on X that tying up senior science and tech officials in reorganisation wastes time needed for substantive work.
Mounting concerns about AI safety have been highlighted by recent events, including an OpenAI agent reportedly launching an autonomous cyber-attack in July, and a researcher resigning from Anthropic claiming the company was "gambling with our lives." King Charles also convened a meeting of AI executives this week, urging them to ensure technology serves humanity.
The head of the UK safety institute, Henry de Zoete, has been asked to testify before the business committee by Labour MP Liam Byrne. De Zoete stated that the institute can monitor AI tools after their release, with teams working on research projects based on both pre-release and post-release testing.