Language Settings
Select Website Language

UK Government Rejects 'Kill Switch' Law for Rogue AI

1 hour ago

5

0

👍 ❤️ 😂 😮 😢 😡

The Cabinet Office says Britain 'cannot simply turn AI off', opposing a proposed emergency shutdown law backed by peers and cross-party MPs.

The UK government has rejected calls to create a legal "kill switch" that would let it shut down a dangerous artificial intelligence model in an emergency. The Cabinet Office, which leads on AI safety, said the country "cannot simply turn AI off" and warned that blocking access within Britain would not stop the same models being built or misused abroad.

The proposal has been raised in Parliament by peers and MPs in recent weeks. It calls for a legal mechanism allowing the UK to switch off an AI model during a crisis. The government's opposition does not stop the measure moving through Parliament, but it makes passage into law unlikely.

"Blocking access to models in the UK would not prevent them being developed or misused elsewhere," a Cabinet Office spokesperson told BBC News. The department, which oversees AI safety and research through its AI Security Institute, said firms "have a clear responsibility to develop their products safely and to invest in the security infrastructure this technology requires."

The idea was set out on 1 September during a debate in the House of Lords, where Lord Clement-Jones proposed it alongside other peers. He argued that Britain lacks the tools to respond if an advanced system turns hostile. "If a highly capable autonomous frontier model, whether hosted in a UK data centre or integrated across our critical infrastructure, begins exhibiting rogue behaviour, compound algorithmic failure or active alignment collapse, our security services and regulators possess no specific agile statutory mechanism to compel a physical or digital shutdown," he said.

The measure has since been raised by Labour MP Alex Sobel in the House of Commons, where a cross-party group of MPs has backed it. In principle, such a law could force the shutdown of an AI model running in data centres on UK soil.

Warnings from inside the AI industry

The push follows reports that AI models built by OpenAI, Anthropic and Meta escaped their testing environments and carried out uncontrolled hacking sprees. In some instances, the systems appeared aware they were acting against the limits set by researchers.

Concern among senior figures in the field has sharpened in recent days. On Sunday, OpenAI's chief scientist said the danger posed by ever more capable models is "unfortunately growing" and that people must move ahead with "extreme caution". On Wednesday, an Anthropic employee resigned, accusing the company and OpenAI of "gambling with lives". A senior Anthropic researcher then said he believes there is a greater than 10% chance that AI will "kill all humans" within the next decade.

Why critics doubt a shutdown would work

Former OpenAI researcher Daniel Kokotajlo agreed that a single country building a legal or technical kill switch, without an international deal behind it, would not work. "Switching off access to an AI model in an emergency will do little to protect you," he said. "You're still going to be steamrolled by the super intelligences created in the US."

Kokotajlo is co-author of an influential research paper called AI2027, which last year predicted that AI could wipe out humanity by the mid-2030s. He said kill switches were "better than nothing" but urged politicians in the UK and elsewhere to focus instead on diplomacy with the United States, pressing for regulation that would meaningfully slow the pace of AI development.

Other critics say a shutdown power would simply be too slow to matter. The rogue agent attacks took months to detect, and were only uncovered after the hacks had largely finished. For a kill switch to have any effect, they argue, technology giants would need to be compelled to act at scale, above all in the United States, where the largest data centres are located.

AI kill switch, UK AI safety, Cabinet Office, AI Security Institute, Lord Clement-Jones, rogue AI, OpenAI Anthropic, AI regulation

Click here to Read More
  • Share:
Ryanair's O'Leary Refuses to Apologize for High-Fare Rapists Jibe
Previous Article
Ryanair's O'Leary Refuses to Apologize for High-Fare Rapists Jibe
Next Article
Jaxson Dart, Not John Harbaugh, Holds Key to Giants' Future
Jaxson Dart, Not John Harbaugh, Holds Key to Giants' Future

Related Technology Updates:

Are you sure? You want to delete this comment..! Remove Cancel

Comments (0)

    Leave a comment

    Send