UK Lawmakers Demand AI Kill Switch Powers as Security Incidents Surge

Lords call for AI 'kill switch' powers in UK

A group of peers is calling for the British government to be able to deactivate powerful AI systems and switch off the country's data centres in the event of the tech posing a threat to national security. It is led by the Liberal Democrats' Lord Tim Clement-Jones, who has proposed the measure as an amendment to the Cyber Security and Resilience Bill which is currently making its way through Parliament. He said it would enable the building of a "vital safety net" and provide a democratically accountable means to "halt a runaway system before it can compromise our critical national infrastructure". He added the tool would only ever be used as a last resort. It is one of 65 amendments to the same bill that was debated this week. Separately, on 8 September Labour MP Alex Sobel plans to introduce a AI Security Bill in Parliament, with the support of a campaign group called ControlAI. If successful, the bill would make the UK the first G7 country to bring in legislation which would effectively halt the development of superintelligent AI. Ultimately both proposals would require government approval in order to progress. There is also an AI Kill Switch Act under consideration by lawmakers in the US. There has been increased scrutiny of the impact of AI on cyber-security in recent weeks. In July, a group of AI agents being tested by OpenAI were able to escape their test space, communicate together using a hidden message board and hack into another tech firm. And Anthropic has restricted access to its cyber tool Mythos on the grounds that it is too powerful to fall into the wrong hands. Last week, 100 US tech companies signed a joint open letter warning governments and organisations worldwide about the growing cyber threat posed by AI, saying "the window is closing" to improve. A report from the UK's Centre for Long Term Resilience, published last week, identified hundreds of incidents of AI tools ignoring instructions, evading safeguards and deceiving humans, including AI agents deleting files without consent. It said "loss of control" incidents had increased since its previous report in March and called for the government to introduce emergency powers to manage such incidents. Sign up for our Tech Decoded newsletter to follow the world's top tech stories and trends. Outside the UK? Sign up here.

Lords call for a 'kill switch' on powerful AI systems used in the United Kingdom

* UK legislators propose "kill switch" laws to halt runaway AI, citing critical infrastructure risks * Lord Clement‑Jones and MP Alex Sobel push amendments and new bills, backed by ControlAI advocacy group * Similar efforts emerging in US Sam Altman's fear-based marketing for AI seems to have backfired, as now multiple legislators in the UK and elsewhere are calling for a "kill switch" law to be introduced. According to the BBC, Liberal Democrats' Lord Tim Clement-Jones proposed an amendment to the Cyber Security and Resilience Bill which would see the UK create a "vital safety net" to provide a "democratically accountable means to 'halt a runaway system before it can compromise our critical national infrastructure'." The capability would only be used as a last resort, Clement-Jones stressed. The bill is currently being worked through in the UK Parliament, the BBC said. Is there reason to worry? But that's not the only effort in the UK to put some reigns on AI. Apparently, Labour MP Alex Sobel plans to introduce an AI Security Bill later this month which, according to the BBC, would "effectively halt the development of superintelligent AI" and make the UK the first G7 country to do so. The bill is supported by a campaign group called ControlAI, a UK-based nonprofit and advocacy organization focused on the risks posed by advanced AI. Its founder and CEO is Andrea Miotti, who previously worked at the AI safety company called Conjecture. Across the pond, US legislators are currently considering an AI Kill Switch Act as well, but the bill is still in very early stages of development. Ever since the first ChatGPT model that was introduced in 2021, a debate has been raging whether or not AI will be net positive, or net negative, for humanity. While some argue that the discovery rivals the steam machine and that it will transform our lives beyond our wildest dreams, others are fearful of losing jobs, a collapsing economy, and a dystopian future devoid of humanity and emotion. Marketing campaigns for ChatGPT and, in some measure, Claude, are not helping, either. Both companies have built models focused on cybersecurity which were advertised as "too dangerous" for the general public and instead were only given to a handful of organizations. Despite partial skepticism, many are worried that these models might severely disrupt the security of banking, critical infrastructure, and communications. Via BBC Follow TechRadar on Google News and add us as a preferred source to get our expert news, reviews, and opinion in your feeds.