Chancellor John Healey has placed artificial intelligence at the center of the coming Budget, setting out commitments worth roughly £600 million in public funding that tie a new biosecurity laboratory to cyber defenses run by AI agents. The measures, announced on 24 October 2026, cover biological threats, hostile attacks and the effect of automation on jobs.
- Key Point: £165 million goes to a National AI Biosecurity Initiative, split between £100 million for a new AI Biosecurity Lab and £65 million for biosurveillance and medical countermeasures against priority pathogens.
- Key Point: OpenAI and Anthropic will supply up to £200 million in model credits, access and technical support for the lab, with Google DeepMind and Basecamp Research named as partners.
- Key Point: A £60 million National Cyber Shield will put agentic AI to work in cyber defense, including a £25 million challenge fund and scoping for a regulatory sandbox.
- Key Point: The AI Security Institute gains £120 million, and £10 million a year will fund research into AI’s effect on jobs and industry.
A laboratory built for biological risk
The largest tranche, £165 million, covers the biosecurity initiative. Within it, £100 million establishes the AI Biosecurity Lab, meant to apply machine learning to detecting and tracking biological threats, while £65 million reinforces biosurveillance and work on countermeasures for priority pathogens. The lab is to be an independent nonprofit rather than a division of a technology company, hiring technical staff from industry and academia and working with public bodies, researchers and frontier model developers. Its remit spans prevention, detection and response, and includes cooperation with the Medicines and Healthcare products Regulatory Agency on health emergency preparation.
Corporate pledges sit alongside the public money. OpenAI has committed £100 million in compute and model credits plus trusted access through its Rosalind Biodefense program, while Anthropic has offered up to £100 million in credits, access and support. DeepMind and Basecamp Research are partners without a published value. The distinction matters: the £200 million is a commercial contribution bound to particular providers’ models and access rules, not Treasury spending, and the terms — eligible workloads, expiry, data handling — remain undisclosed.
“Our new AI Biosecurity Lab will be a major new global institution, bringing together world-leading scientific expertise to protect people from emerging biological threats and ensure Britain leads the world in AI safety and security,” AI Minister Kanishka Narayan said.
Agents on the cyber front line
The £60 million National Cyber Shield is built as a deployment program for AI agents rather than a conventional procurement. It carries a £25 million challenge fund for British AI and cyber security innovators and scoping work for a sandbox where agentic tools could be tested across critical national infrastructure, the public sector and industry. A further £70 million goes to a National Exposure and Real-Time Vulnerability Centre, which would turn AI loose on weaknesses in critical systems. Existing blueprints describe agents that find threats at machine speed, share evidence and eventually assist with remediation, while staying under their owners’ control; early deployments target government and critical sectors.
The backdrop is concrete. The UK government was among those that took some sites offline as a precaution after researchers exposed serious flaws in PHP, a server-side scripting language long used in web development. No successful exploitation of the flaw, nicknamed php2shell, has been reported, but specialists warned the episode shows how quickly dormant bugs can surface — and be attacked — with AI.
Chips, jobs and the test of delivery
Outside security, the package funds AI chip design capacity in Edinburgh and an AI Centre of Excellence in Manchester, plus £10 million a year for work on how AI will reshape jobs. The AI Security Institute, created in 2023, gains £120 million to keep testing advanced models and developing technical standards.
“I understand and share the concerns of working people, so I am setting out a distinctively British approach to AI — one that doesn’t leave workers behind and defends our national interest, while ensuring the benefits of AI-enabled growth are felt right across the country,” Healey said.
Security Minister Dan Jarvis said the funding would help Britain “seize the opportunities AI presents to strengthen our national security, while staying one step ahead of those who seek to use AI to do us harm.”
What remains unresolved is execution. The figures describe planned investment, not money contracted or spent, and no shared delivery timetable has been issued. Procurement schedules, milestones, audit requirements and outcome measures are still missing for most of the package, as are the credit terms and how researchers will compare providers while protecting sensitive biological data. The Budget will be the first test of how much of the £600 million becomes institutions rather than announcements.