UK Foreign Secretary Warns of ‘AI Hiroshima’ if Policymakers Don’t Act

Governments risk repeating mistakes made at the dawn of the nuclear age if they wait to legislate for artificial intelligence, British Foreign Secretary Yvette Cooper has warned.

In one article Published Monday, Cooper said AI offers breakthroughs but also poses new risks as the technology becomes more powerful and widely available.

“Last month in Shenzhen, China, I saw the extraordinary promise of AI and robotics being used for life-saving healthcare,” Cooper wrote. “But these same technologies are also reshaping the future of warfare, crime and social cohesion in alarming ways.”

Cooper said managing the risks of AI could become “the biggest security challenge of the next decade.” He argued that governments need international agreements on cutting-edge technology before a crisis forces action.

She compared the current race to develop AI systems to the early nuclear arms race, saying global security agreements emerged only after countries witnessed the devastation caused by nuclear weapons.

“An international agreement on nuclear energy was only reached after the world saw the terrifying power of the new technology in Hiroshima,” Cooper wrote. “We cannot afford to wait for an AI equivalent of Hiroshima before we take action.”

Cooper called on Britain to use its diplomatic influence to bring together the United States, China and other major AI powers to establish shared safety principles and standards, citing the 2023 AI Safety Summit in Bletchley Park, which brought together world leaders from 29 countries and the European Union to discuss the emerging risk of AI. Cooper called this an example of Britain’s ability to “unify the world on AI security.”

Cooper’s warning comes after months of escalating concerns about how governments should oversee increasingly powerful AI systems.

In May, the UK AI Security Institute warned of rapid gains in AI cybersecurity capabilities after OpenAI’s GPT-5.5 became the second model to complete a simulated cyber attack without human assistance, following Anthropic’s Claude Mythos Preview.

Days later, the International Monetary Fund warned that AI could “amplify” cyber attacks against the global financial system by reducing the skills needed to exploit vulnerabilities, and urged policymakers to treat cybersecurity as a matter of financial stability rather than a purely technical problem.

In June, President Donald Trump signed an executive order that created a voluntary framework for reviewing cutting-edge AI models before release, expanding AI cybersecurity programs and directing agencies to evaluate potential national security risks from cutting-edge AI models.

That same month, calls for stricter regulations also emerged from the AI ​​industry, with Anthropic CEO Dario Amodei arguing that transparency requirements are no longer enough and calling for mandatory third-party testing of borderline models. This demand was followed by the US government ordering Anthropic to restrict access to Claude Fable 5 and Mythos 5 due to national security concerns, before the order was lifted in July.