1,000+ Opportunities
Find the right grant
Search federal, foundation, and corporate grants with AI — or browse by agency, topic, and state.
Frontier AI Discovery is the UK's principal attempt to move domestic frontier AI capability out of the lab and into deployable technology, and its two-phase structure is the single most important thing to understand before applying. Phase 1 is deliberately small and fast - GBP 25,000 to GBP 50,000 in eligible costs, funded at 80 percent FEC, over a project of no more than three months.
It exists to buy an option, not to fund a research programme. What it buys is eligibility for Phase 2, where awards jump to between GBP 5,000,000 and GBP 10,000,000 delivered through collaborative consortia. Applicants who treat Phase 1 as a standalone feasibility grant are misreading the competition; the assessment is effectively asking whether this team and this idea could plausibly absorb an eight-figure follow-on award.
Scope covers frontier artificial intelligence, machine learning and foundation models, with the competition documentation pointing at application domains including healthcare, advanced materials and national security. Phase 1 is open to single applicants only - no consortium is required or permitted at this stage - and any funded organisation must carry out project work in the UK and intend to exploit the results from or in the UK.
The 2026 Phase 1 round opened in spring and closed on 10 June 2026 at 11:00am UK time, with funded projects running from 1 October 2026 to 31 January 2027, meaning the current cohort is now in delivery and Phase 2 is the live opportunity for those who came through it.
Organisations that missed Phase 1 should monitor the Innovation Funding Service and UKRI opportunity pages for the next cycle, since DSIT has signalled continued frontier AI investment and the programme sits alongside the separate AI Champions and Sovereign AI streams rather than replacing them.
Get a weekly digest of new grants like this
A free weekly digest of new foundation and federal funding opportunities as they're added to Granted. Unsubscribe anytime.
Or search similar grants →According to the current listing, eligibility includes: Phase 1 is open to single applicants only; collaborative applications are not accepted at this stage. The applicant must be a UK registered organisation, must carry out all project work in the UK, and must intend to exploit the project results from or in the UK. Total eligible project costs for Phase 1 must fall between GBP 25,000 and GBP 50,000, with research organisations funded at 80 percent of full economic cost and business funding rates determined by organisation size under standard Innovate UK subsidy rules. Projects must last no more than three months. The 2026 cohort was required to start on 1 October 2026 and finish by 31 January 2027. Technical scope is frontier AI, machine learning and foundation models with a credible and defensible route to scale-up; the competition explicitly welcomes applications targeting healthcare, advanced materials and national security among other sectors. Applications are submitted through the Innovation Funding Service; the 2026 Phase 1 deadline was 10 June 2026 at 11:00am UK time and that round is now closed. Completion of a Phase 1 project is the gateway to Phase 2, which awards GBP 5,000,000 to GBP 10,000,000 to collaborative consortia - applicants should therefore be able to describe, at Phase 1, the consortium and scale-up pathway they would assemble for Phase 2. Universities should note that some institutions run internal selection ahead of submission, so internal deadlines can fall six weeks or more before the Innovate UK deadline. Confirm the full requirements in the official notice before applying.
The current listing shows phase 1 feasibility studies carry total eligible project costs of GBP 25,000 to GBP 50,000 (approximately USD 32,000 to USD 65,000), funded at 80 percent of full economic cost for research organisations, with projects lasting up to three months. Phase 1 is a gateway rather than an endpoint: successful Phase 1 applicants become eligible to apply for Phase 2 awards of GBP 5,000,000 to GBP 10,000,000 (approximately USD 6,500,000 to USD 13,000,000) delivered by collaborative consortia. Verify award ceilings, matching requirements, and allowable costs in the official notice.
Innovate UK Frontier AI Discovery Programme for Foundation Model and Machine Learning Feasibility Studies is funded by Innovate UK (part of UK Research and Innovation, UKRI), delivered with the UK Department for Science, Innovation and Technology. Verify program details on the funder's official page before applying.
Start from the official opportunity page linked in this listing — it carries the sponsor's submission instructions.
On August 26, 2026, NSF named the STC class of 2026: TEMPEST at Michigan State, GENIE at Northwestern, and a human-robot co-adaptation center at UT Austin. $90 million total, $6 million a year each. The 2023 class was four centers and $120 million. The award size is intact; the shelf space shrank 25 percent — and that is the number that should reshape how you plan a center proposal.
Read articleOn July 22, 2026 the White House committed more than $5 billion to expand the Genesis Mission — the government-wide 'AI for science' effort — announcing 278 first projects and 14 National Science and Technology Challenges spanning health, energy, infrastructure, manufacturing, and national security. More than 15 federal agencies are contributing awards, datasets, and facilities. Here is how the pieces fit, why it reorganizes the science-funding map, and how to position for the awards flowing out of it.
Read articleOn July 22, 2026 the White House announced more than $5 billion and 15-plus federal agencies behind the Genesis Mission — a national effort to point AI at the hardest problems in science. Here is what the mission actually funds, how its five National Science and Technology Challenges map to real solicitations already opening, and how a research team or small business should position for the money flowing out of it.
Read article