Haigh urges UK to take AI safety warnings seriously

Louise Haigh said the UK must heed warnings from AI industry leaders, arguing that the technology could improve public services, strengthen businesses and drive scientific breakthroughs, but also carries huge risks to national security and society without proper safeguards.

Labour MPs and peers are urging ministers to work with international partners on regulation. Darren Jones called for Andy Burnham to put international G20 regulation on the agenda when the UK hosts the summit next year, while TUC general secretary Paul Nowak said the technology posed a clear and present danger to workers and wider society.

Liam Byrne has asked AI Safety Institute head Henry de Zoete to testify to parliament amid concerns about unfettered and inadequately governed development. Anthropic will not face sanctions after declining to submit its latest model for testing before launch, even as the company has called for a coordinated global slowdown backed by rivals including OpenAI, Google DeepMind and Elon Musk.

Union delegates in Brighton warned that workplace AI is already being used to monitor teachers, crane operators, HGV drivers and office workers. Haigh said working people should be central to the debate, and that productivity gains should benefit workers rather than only shareholders or tech executives.