- Published on
UK must heed warnings from AI experts
- Authors
- Name
- DeepHorus
Louise Haigh, the UK's first secretary, will on Tuesday urge ministers to “heed the warnings” from AI industry leaders about the existential threats posed by the technology, even as the government seeks to capitalise on its benefits. In a speech to the TUC congress, she will vow that the government will take public safety and national security concerns seriously in partnership with international allies. The address comes amid escalating alarm from the very labs building advanced AI systems.
Over the weekend, Anthropic urged its competitors and governments to coordinate a global slowdown in AI development, a call backed by OpenAI, Google DeepMind, and Elon Musk's X. Three Anthropic researchers had earlier warned that artificial intelligence could kill off humanity within the decade. Labour MPs and peers have now joined the chorus, demanding deeper international cooperation on regulation.
Haigh will tell trade union delegates that AI has “enormous potential to transform our public services, make our businesses stronger and deliver new scientific breakthroughs.” However, she will stress that the government will look seriously at public safety and national security threats with international partners. Her speech will also outline priorities for public control of transport, water, and energy.
On Monday night, Anthropic co-founder Jack Clark suggested that a third-party “kill switch” for AI systems might need to become mandatory for companies. He said society “might want to eventually pass rules around” such a mechanism. This would represent a significant shift from voluntary self-regulation toward binding safety standards for the most advanced models.
The immediate impact is on the UK's approach to AI governance, as ministers balance economic opportunity against existential risk. Trade unions, a key Labour constituency, are being positioned as stakeholders in the safe deployment of AI in public services. Meanwhile, companies like OpenAI are pushing for narrowly targeted legislation that would affect only the most powerful AI labs, not smaller startups.
OpenAI's head of policy in Europe, Tom Duff Gordon, said: “We support stronger UK rules for the handful of companies, including OpenAI, developing the most powerful AI systems.” However, he cautioned that legislation should be narrowly focused on advanced labs and not burden less powerful startups. The mixed response highlights the industry's internal divisions over how to regulate a technology that many of its own creators fear.
Haigh's speech is expected to signal a more proactive UK stance on AI safety, potentially leading to new legislation in the coming months. The government's willingness to collaborate with international partners will be tested as it seeks to balance sovereignty, innovation, and security. A global slowdown in AI development, as urged by Anthropic, remains an open question.
This moment echoes previous technological inflection points, where governments rushed to catch up with transformative innovations. But the unique, existential nature of AI risk has pushed the debate beyond typical regulatory concerns into questions of human survival. With the TUC congress as her backdrop, Haigh is attempting to frame AI as a tool for public good, while acknowledging the warnings of those who build it.
