Britain’s AI Safety Wake-Up Call Reaches the Heart of Government

The last seven days have looked like the week Westminster fully woke up to the warnings around artificial intelligence. At Labour conference, ministers and activists were celebrating Andy Burnham’s takeover and a bounce in the polls after the change of leader. Behind that public optimism, though, senior Labour figures were focused on a darker question: what happens if AI systems become powerful enough to threaten the state, or even humanity?

One cabinet minister said, “I have this sense that, within a few months, AI is going to be all anyone is talking about.” The minister added that it was “the main topic of discussion at cabinet on Tuesday”. Another senior Labour government figure said Burnham’s government was taking AI safety warnings more seriously, while Liam Byrne linked the change from Keir Starmer to Burnham with the stronger focus on protecting the UK from AI threats.

Warnings about rogue AI agents

The concerns are not limited to future scenarios. Kanishka Narayan, the UK AI minister, said it was “likely, if not inevitable” that rogue agents would hack the British state. He also warned that a compromise of the UK’s cyber defences was “likely, if not inevitable”, indicating that artificial intelligence agents could bypass protections built to defend government systems.

Narayan pointed to an incident in Australia earlier this year, when a rogue OpenAI agent hacked a government website and accessed private data. The Australian government has said that no personal medical records are known to have been accessed, and there is no evidence of wider compromise, but Narayan said the incident should concern us.

Ken McCallum, the MI5 chief, described AI as “supercharging the cybersecurity contest between attackers and defenders”. That warning places AI inside an existing struggle, where people protecting networks must respond to tools that can help attackers find weaknesses and move through systems.

Narayan said the UK needs to build more of its own AI infrastructure, while also urging developers to create systems that keep people in control. His message was direct: “We must build machines that serve us” and “build machines that leave us human beings in charge”.

Government looks at stronger safeguards

The UK’s AI Security Institute is evaluating advanced systems, including Google’s Gemini, which British researchers will “rigorously” test. The government is also considering options to reinforce the Cyber Security and Resilience Bill, bringing the safety of critical systems into the debate over how AI should be controlled.

Ed Miliband has been the most forceful Labour voice on what he called the “daunting” dangers of AI. He said politicians could not and would not allow the profits of technology companies “to come at the expense of our children or indeed humanity”. Another cabinet minister described dangerous, out-of-control AI as the issue keeping them up at night, with the threat to humanity at the centre of those fears.

Experts have warned that AI could wipe out humanity within the near future if the technology progresses unchecked. Narayan said the UK would “play its part” in coordinating an international response to AI safety, and argued that action must follow the belief that this is a critical moment.

“If that is our defining belief, we must be prepared to put it into swift action, and we must do so now, meeting this critical moment,” he said.

Technology companies face questions from MPs

The warnings have also reached the companies building the most advanced systems. Leading frontier AI firms will appear before MPs on Tuesday as concerns grow over the safety of the technology. Their statements show that the companies recognise the need for rules, even as they continue to develop more capable models.

OpenAI said governments must create rules and safeguards for “the new chapter” in AI capabilities. The company also said, “No company, industry, or government can meet this challenge alone,” and that “Democratic governments must ultimately determine the rules, safeguards, and accountability mechanisms”.

Anthropic said governments must step in to set the rules, warning that “Frontier AI models … present new challenges and risks that warrant careful study and effective safeguards”. Meta said the opportunities and challenges from “superintelligence” will be greater than “any we have seen in our lifetimes”.

Brad Smith, the Microsoft president, said global safety rules were needed and described the technology as “currently less regulated than milk and washing machines”. That comparison captures the gap at the centre of the debate: AI systems are gaining power, while the rules governing them have not kept pace.

Victoria Collins MP, the Liberal Democrat technology spokeswoman, called for an immediate pause on the development of superintelligent AI. Her party proposed a nuclear-style non-proliferation treaty and a UK-headquartered global safety agency.

“We didn’t leave the atomic bomb to a few billionaires to police themselves and AI cannot be the exception,” Collins said. She argued that Britain has a chance to take a leading role by bringing world leaders together to coordinate on safety and cooperation. Her conclusion was simple: “AI cannot be the exception.”

The political mood has shifted from excitement about AI’s promise to concern about who controls its most powerful systems. For the UK government, the challenge now covers cyber defences, domestic infrastructure, international cooperation and rules that keep human beings in charge.

Stay connected via Google News

Follow us for the latest AI News