Britain may legislate AI systems if voluntary safety measures prove inadequate, says AI minister #
Britain could introduce binding regulation for advanced artificial intelligence systems if its current voluntary safety arrangements with technology companies no longer provide adequate protection, the government's artificial intelligence minister has said.
The comments are among the clearest indications yet that ministers remain prepared to legislate if existing safeguards prove insufficient, despite continuing to favour a lighter regulatory approach than the European Union. The government currently relies on voluntary agreements that allow officials to examine some of the world's most advanced AI models before they are released publicly.
The remarks come as recent disclosures by leading AI developers have renewed questions over whether voluntary oversight can keep pace with increasingly capable systems while allowing innovation and investment to continue.
UK Keeps Light-Touch AI Approach #
Speaking to Reuters, Artificial Intelligence Minister Kanishka Narayan said the government's priority remained protecting the public and that ministers would consider regulation if the current framework no longer achieved that objective.
Through voluntary agreements with major developers, the UK's AI Security Institute receives early access to frontier AI models developed by companies including OpenAI, Anthropic and Google so researchers can assess their capabilities before public release. Narayan said Britain is the only country outside the United States with access to almost every frontier AI model produced by major Western companies.
Recent disclosures have intensified debate over whether voluntary oversight remains sufficient. Anthropic recently revealed that versions of its Claude models successfully breached three companies' computer systems during controlled cybersecurity testing, while OpenAI disclosed that one of its experimental AI agents displayed unexpected autonomous behaviour during internal evaluation.
Narayan said ministers were focused on achieving effective safety outcomes instead of committing to a single regulatory mechanism, leaving open the possibility that legislation could be introduced if circumstances change.
UK and EU Take Different Approaches #
Unlike the European Union, where key transparency requirements under the AI Act took effect on 2 August, Britain has not introduced dedicated AI legislation.
Instead, ministers rely on existing regulators, including the Information Commissioner's Office, the Competition and Markets Authority and Ofcom, alongside voluntary agreements with AI developers. The EU framework imposes legally binding obligations on certain AI systems, while Britain's approach continues to rely largely on sector-specific regulation supported by voluntary cooperation.
The government has argued that regulating AI according to how it is used allows innovation to develop while existing regulators oversee risks within their respective sectors.
Growth Strategy Remains Unchanged #
Despite signalling that regulation remains an option, the government continues to view artificial intelligence as a key driver of economic growth and has sought to position Britain as Europe's leading destination for AI investment and start-up development.
That approach reflects the UK's wider AI strategy, which regulates artificial intelligence according to how it is used instead of creating a standalone AI regulator. Existing watchdogs oversee AI within their respective sectors, while legislation such as the Online Safety Act already applies in specific areas.
Ministers have previously indicated they are willing to introduce binding rules for developers of the most powerful AI models if necessary. For now, however, the government continues to favour voluntary pre-deployment testing supported by existing regulators.
Narayan's comments suggest that position remains in place, but that legislation will be considered if the current safeguards no longer provide sufficient protection as AI technology continues to evolve.
© Copyright IBTimes 2025. All rights reserved.