Backed by £59m, ARIA’s Safeguarded AI programme sits within the Mathematics for Safe AI opportunity space and aims to demonstrate a world where we can use fleets of AI agents to model and verify critical cyber and cyber-physical systems, radically boosting societal resilience.
As part of our updated programme thesis, we seek to test and accelerate the hypothesis that AI-enabled formal methods can make high-assurance cyber defence practical at scale.
Within Technical Area 2, this funding call pursues the question: given the advances in AI and formal methods, what are the most ambitious, security-critical systems we can verify today?
We will fund teams to build production-grade, security-critical software components whose key security properties are backed by machine-checked proofs, validated through coordinated red-team exercises.
We will evaluate teams for their credible interest and ability to carry developed capabilities towards large-scale deployment, whether themselves or through spinouts, partnerships, upstreaming, or other routes.
We’re accepting proposals for Blue Teams to build production-grade, security-critical software components whose key security properties are backed by machine-checked proofs, under clearly stated threat models and assumptions – using AI as a central means of reaching levels of ambition for formal proof that would otherwise not be tractable. Funded teams will work in 8-week sprint cycles, with each verified system stress-tested by a central red-teaming partner, and will be supported toward real-world adoption.
Applications may be submitted at any time while the call is open, and are reviewed in quarterly batches.
Key dates for the current cycle:
Applications open: 4 September 2026
Proposal cut-off: 31 October 2026
Review period: 1–30 November 2026
Applicant notification: 30 November 2026
Proposals submitted after a cut-off will roll into the following quarterly cycle. ARIA expects to fund a small number of projects through this route and may choose not to make awards in a given review cycle. We may update cut-off dates, pause the call, or close it early if needed.
We welcome applications from across the R&D ecosystem, including startups, industry and academia. We typically require the majority of the project work to be conducted in the UK (i.e. >50% of project costs and personnel time).
For further details, read the call for proposals.
Join a team
Applicants looking for collaborators with complementary expertise across target domains, security, formal methods, and AI can submit their information to our teaming platform and identify others they may wish to reach out to.
-
We welcome proposals from across the R&D ecosystem – single organisations or teams spanning academia, industry, and non-profits. Strong teams are likely to combine deep expertise in security and systems engineering, formal methods and proof engineering, and AI-assisted workflows, with the ability to iterate quickly in short, evidence-led cycles and a credible orientation toward real-world deployment.
Applicants looking for collaborators with complementary expertise across target domains, security, formal methods, and AI can submit their information to our teaming platform and identify others they may wish to reach out to.