The $5.5tn (£4.1tn) US chipmaker Nvidia is donating $250m worth of computing power to the UK to support AI safety research.
The donation from the world's most valuable company comes as the UK government confirmed its AI Security Institute (AISI), which tests advance AI models, will be able to examine cutting-edge technology developed by companies such as OpenAI, Anthropic and Google before they are released.
Nvidia said the donation, which spans five years, would "support agentic safety and security research across the UK".
Safety concerns about AI agents – highly capable computer programs that are able to design and perform sequences of tasks autonomously – have risen since the summer, when OpenAI confirmed that a "swarm" of agents had attacked Hugging Face, a tech startup.
Nvidia did not reference AISI in its statement but said most of the money would "support the wider UK AI safety and security research community", with a proportion of the cash going to the British Open-Ended Learning and Discovery Lab, the Science of Fundamental AI Research Lab – run by the Edinburgh, Cambridge, Oxford and University College London universities – and Imperial College London.
The computing power will be delivered via UK-based datacentres using Nvidia chips.
"The UK brings world-class scientific rigour to this essential engineering challenge," said Jensen Huang, Nvidia's chief executive. "With this $250m commitment, Nvidia will put accelerated computing in the hands of UK researchers to develop, test and strengthen safeguards for increasingly powerful AI."
Kanishka Narayan, the UK's AI minister, said the announcement was a "vote of confidence in our world-leading strength and expertise in AI research".
On Monday Narayan confirmed that AISI was conducting tests on US-developed AI models before they are released, following reports that the White House had restricted access.
The AI minister also warned that the UK faces agentic attacks similar to the one carried out by OpenAI agents on Australia's health care system this year.
"It seems likely, if not inevitable, that our own cyber defences will be breached," said Narayan.
Anthropic, the developer of the Claude chatbot, warned of a new AI safety threat on Monday after it published a blogpost stating its models could soon become self-replicating. They find ways to infiltrate internet-connected devices such as smart ovens and lawnmowers, the company said.
Under such a scenario the models could power themselves without datacentres, and instead could use a "botnet" of interconnected devices, Anthropic said, adding that the world should prepare for a future where such botnets were here to stay.
"A replicator like this would be challenging to shut down since there's no centralised point to shut off," Anthropic's researchers wrote.