1. Home
2. News
3. AI safety

Nvidia pledges $250 million in computing over five years for UK research on AI agent safety

The chips come through British cloud partners, and Nvidia hasn't picked them yet.

By Think Facility · October 14, 2026 · Nvidia AI safety Save

Nvidia said on October 14, 2026 that it will give British researchers $250 million worth of computing over five years, and all of it is meant for work on keeping AI agents safe and secure. That comes to $50 million in compute access a year. The announcement calls it a donation.

Most of it goes to the wider UK research community working on AI safety and security. A share is set aside for Oxford's BOLD lab, UCL's SOFAIR lab and Imperial College London.

Who

NVIDIA

Value

$250 million over five years

Each year

$50 million in compute access

Named partners

BOLD, SOFAIR and Imperial College London

Delivered through

U.K.-based NVIDIA Cloud Partners

What Huang said

Jensen Huang, Nvidia's founder and CEO, gave the reason in one line, and the rest of his statement is about what British researchers will do with the chips.

He says the money will let them "develop, test and strengthen safeguards for increasingly powerful AI." None of it arrives as cash, though. Nvidia will pay for access through UK-based cloud partners, with the hardware hosted in Britain, and the release says provider selection, technical architecture and commercial terms are still to be agreed. So the researchers don't yet know whose data center they'll be logging into.

What the research covers

The programs center on secure agents, the security of open models and defenses against real attacks. Some of the work will run in secure sandboxes and build benchmarks for judging how safe a model is. Nvidia also names its own tools: the OpenShell runtime and the Sentry reference design, the two halves of the agent safety platform it launched on September 28, 2026.

The lab heads each got a line. Jakob Foerster, BOLD's founding director, said "AI safety as an afterthought is no longer appropriate for the open-ended, evolving world." Imperial's Alessandra Russo pointed at robots instead.

What the government said

Kanishka Narayan, the UK's AI minister, called the pledge "a vote of confidence in our world-leading strength and expertise in AI research." Two days earlier he'd given a speech telling the labs not to release systems they can't control. This time he tied the money to the UK's G20 presidency, where he said a safe and secure British AI future will be a central objective.

Business Secretary Jonathan Reynolds said people need to feel the technology is safe before the country can use it. We couldn't find a government copy of the announcement, so both ministers' words here come from Nvidia's release.

It's the second pledge of its kind from Nvidia in a week. On October 8, 2026 it announced $1 billion over five years to advance US science.