Government’s Firm Position on a Kill Switch
On Friday, the Cabinet Office issued a clear rebuttal to mounting demands for a “kill switch” that would allow the state to shut down an advanced artificial intelligence system in an emergency. In a formal statement, the office declared that Britain “cannot simply turn AI off” and argued that restricting access to models within the United Kingdom would not stop them from being created or misused elsewhere. The spokesperson emphasized that companies bear a clear responsibility to develop their products safely and to invest in the security infrastructure required for responsible AI deployment.
“Britain cannot simply turn AI off and blocking access to models in the UK would not prevent them being developed or misused elsewhere,” the statement read, underscoring the administration’s refusal to grant the government a last‑resort power to disable autonomous models during a crisis.
Parliamentary Push for Emergency Powers
Amid intensifying concerns, several members of Parliament have urged the introduction of statutory safeguards that would enable the state to intervene if an AI system malfunctions or is weaponised. Liberal Democrat peer Lord Clement‑Jones tabled an amendment to the Cyber Security and Resilience Bill, proposing a legal mechanism for emergency shutdowns. He argued that such a provision would provide “a necessary emergency kill switch” when the Government “currently lack clear statutory power to direct an emergency shutdown”.
While the amendment has yet to pass, the debate reflects a broader parliamentary movement to embed robust oversight into the rapidly evolving AI landscape, as legislators seek to pre‑empt potential disasters before they materialise.
Industry Alarm and Expert Warnings
The Prime Minister recently acknowledged that AI poses “risks to national security”, a comment made in the wake of a stark warning from a leading safety researcher at Anthropic, who claimed the technology could “kill all humans” within the next decade. Evan Hubinger, alignment science lead at Anthropic, posted on social media that he “personally think[s] it is >10% within the next decade” that AI could cause human extinction, adding that the company “does not yet have a plan to solve alignment for superintelligence”.
Jacob Coxon, another Anthropic researcher who resigned from the firm, warned that a competitive race toward self‑improving superintelligence amounted to “gambling with our lives”. In his resignation statement, he asserted that “the people building AI earnestly believe that it could kill us all by the end of the decade. This is not a marketing stunt.” He further criticised both Anthropic and its main rival, OpenAI, for acting “irresponsibly” and noted that “no other human activity poses this level of danger.”
According to the Financial Times, the US technology giant has withheld its latest model, Claude Mythos 5.1, from the UK’s AI Safety Institute, making it available only to vetted American organisations. This move has heightened anxiety among UK safety officials who rely on independent testing to assess the technology’s hazards.
Policy Direction and Future Safeguards
In response to the escalating discourse, Andy Burnham has elevated the role of AI safety minister Kanishka Narayan by granting him a seat at the Cabinet table. Burnham said the appointment had “lifted the level of the conversation about artificial intelligence within Government” and stressed that “the offer is open to colleagues on all sides of the House to be part of that crucial discussion”.
Officials have indicated that while the Government remains opposed to a blanket kill switch, it is exploring “targeted interventions” that would address the most significant risks linked to AI. The Cyber Security and Resilience Bill is expected to bolster UK cyber defences and improve the security of essential public and digital services, including data centres. As the debate continues, the administration insists it will adopt a “long-term, science‑led approach” to understand and prepare for emerging AI threats, working closely with its world‑leading AI Security Institute.
Why It Matters
The UK’s refusal to adopt a kill switch, juxtaposed with growing parliamentary and expert pressure, signals a decisive moment in the nation’s approach to AI governance. By rejecting an outright shutdown mechanism, the government risks appearing complacent in the face of existential warnings, while simultaneously signalling a willingness to pursue nuanced, proportionate safeguards. The outcome will shape not only domestic policy but also the United Kingdom’s credibility on the global stage as a leader in responsible AI development and regulation.
