UK Sets Out £600 Million AI Security Package, With OpenAI and Anthropic Backing New Biosecurity Lab

The British government on Oct. 24 detailed a package of more than £600 million for artificial intelligence that puts security at its center, led by a new AI Biosecurity Lab backed by up to £200 million in model and compute credits from OpenAI and Anthropic. HM Treasury said Chancellor John Healey will set out the measures as part of his Budget next week, alongside new money for the AI Security Institute, two AI-driven cyber defense programs and a state-backed fund for British AI companies.

Policy background

The United Kingdom has tried to stake out a role as a convener on AI safety since it hosted the first global AI Safety Summit at Bletchley Park in 2023 and created what is now the AI Security Institute. The new package extends that strategy from evaluating frontier models toward applying them, particularly in biology, where officials see both the largest upside and the most serious misuse risks.

According to the Cabinet Office, the biosecurity lab will be modeled in part on the AI Security Institute and will work with the UK Health Security Agency, the Animal and Plant Health Agency, the Defence Science and Technology Laboratory and the Medicines and Healthcare products Regulatory Agency. Its first priorities are early detection of emerging biological threats and faster development of vaccines, treatments and other countermeasures. Google DeepMind will offer research collaboration and intends to provide trusted access to Gemini and specialized science models, while London-based Basecamp Research will contribute access to its AI biology models. The government did not give a headquarters location, launch date or governance timetable and said details of a scientific advisory committee would follow.

The Treasury framed the wider package around three goals: security, homegrown capability and growth across regions. It also set aside £10 million a year to study how AI will reshape jobs, £23 million over three years for an AI Economics Institute, a £15 million AI Challenge Prize with a first award aimed at health, and £30 million for local AI partnerships with mayoral authorities.

Industry implications

For AI developers, the structure of the biosecurity deal is the notable detail. OpenAI and Anthropic are contributing credits and trusted model access rather than cash, which ties a government research body directly to their platforms. The Cabinet Office said OpenAI's contribution includes trusted access through its Rosalind Biodefense program. That gives both companies a formal role in national biodefense work while giving the lab access to frontier systems that would be costly to replicate.

The cyber measures point to growing government demand for agentic security tools. The National Cyber Shield includes scoping of a regulatory sandbox for agentic cyber defense across critical national infrastructure, the public sector and industry. If the sandbox moves forward, it could become a route for British security startups to test autonomous tools with operators of energy, water, telecom and transport networks.

The £120 million top-up for the AI Security Institute will likely deepen pre-deployment testing of frontier models, an area where developers increasingly work with government evaluators before release.

Competitive and supply chain implications

The package tilts toward capability the UK can control at home. The Sovereign AI Fund will invest directly in domestic AI companies and add robotics, autonomous systems and advanced manufacturing as a focus area. The £60 million procurement scheme covering health and care, AI security and assurance gives early-stage firms a potential first public customer, often the hardest step for deep-tech suppliers.

The £18 million chip design academy targets one of the UK's clearest semiconductor strengths. Britain has limited leading-edge fabrication but a strong base in chip architecture and design. The Treasury linked the academy to advanced semiconductor manufacturing and reindustrialization, and named Edinburgh as a location for chip design activity, although it gave no separate funding figure. Training more design engineers could ease talent constraints for UK chip designers and for global companies with British design centers.

The biosecurity lab could also shape demand for diagnostics, genomic surveillance and sequencing. Oxford Nanopore Technologies was among the organizations that welcomed the announcement, and the lab's emphasis on early pathogen detection lines up with that segment of the UK life sciences supply chain.

Regional implications

Manchester gains two institutions: an AI Centre of Excellence and a National Cyber Security Centre AI Cyber Security Centre of Excellence to train government cyber and AI specialists. Together with Edinburgh's chip design role and the £30 million for mayoral authorities, the plan spreads AI investment beyond London and the Oxford-Cambridge corridor.

Internationally, the biosecurity lab is pitched as a global institution, and the Treasury tied AI Economics Institute research to the UK's G20 presidency next year.

What comes next

• Healey is expected to confirm the package in his Budget next week.
• The government is due to publish governance details and the scientific advisory committee for the AI Biosecurity Lab.
• The National Cyber Shield's £25 million challenge fund and the scoping of an agentic cyber sandbox should open to UK innovators.
• The Sovereign AI Fund will begin investing under its new physical and industrial AI focus.
• The AI Challenge Prize is set to make its first award, aimed at health applications.

Sources

• HM Treasury: Chancellor unveils Britain's ambitious plan for AI (Oct. 24, 2026)
• Cabinet Office: Introducing the AI Biosecurity Lab (Oct. 24, 2026)

Image: HM Treasury building on Whitehall, London, by Tilman2007 via Wikimedia Commons, CC BY-SA 4.0. Cropped and edited by Credence Research.