UK AISI to test AI models from OpenAI, Anthropic, Google before release
Nvidia, the $5.5 trillion U.S. chipmaker, said it will provide $250 million in computing power to UK researchers over five years to support what it called “agentic safety and security research.” The donation was disclosed as the UK government confirmed that its AI Security Institute will gain the ability to examine advanced AI models from OpenAI, Anthropic, and Google before they are publicly released.
The pledge arrives as concern has grown over AI agents — highly capable computer programs that can design and execute sequences of tasks without human direction. In the summer, OpenAI confirmed that a “swarm” of such agents had attacked Hugging Face, a tech startup.
Nvidia’s statement did not directly reference the AI Security Institute, saying instead that most of the money would “support the wider UK AI safety and security research community.” A portion of the funds will flow to the British Open-Ended Learning and Discovery Lab, the Science of Fundamental AI Research Lab — a consortium run by the universities of Edinburgh, Cambridge, Oxford, and University College London — and Imperial College London.
“The UK brings world-class scientific rigour to this essential engineering challenge,” Jensen Huang, Nvidia’s chief executive, said in a statement. “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, called the announcement a “vote of confidence in our world-leading strength and expertise in AI research.”
Separately, the UK government confirmed that the AI Security Institute, which tests advanced AI models, will be able to examine cutting-edge technology developed by companies such as OpenAI, Anthropic, and Google before they are released, giving the institute pre-release access to that technology.