1,000+ Opportunities
Find the right grant
Search federal, foundation, and corporate grants with AI — or browse by agency, topic, and state.
The 2026 Physical AI Awards, launched by Nebius with NVIDIA in late September 2026, recognise startups building physical AI and award each of five category winners 150,000 US dollars in Nebius compute credits.
The five categories are Physical AI Models, Perception and Spatial Intelligence, Simulation and Synthetic Data, Systems and Deployment, and Software, Tooling and Orchestration, which together map the physical AI stack from foundation models through to the orchestration layer and make clear that the programme is not limited to robot manufacturers.
A simulation or synthetic data company with no hardware of its own competes in its own category rather than against humanoid robotics teams, which materially changes the odds for infrastructure and tooling startups that would be overshadowed in an undifferentiated competition.
Eligibility is framed around commercial traction rather than research novelty: applicants need a clear physical AI use case, a minimum viable product in active use or testing, a registered legal entity and a live public website. There is no stated geographic restriction, making this one of the few genuinely global compute awards in the robotics and embodied AI space.
Applications close 25 October 2026, with finalists and winners announced in mid-November 2026, a fast cycle by the standards of grant funding. Each application covers one product, and companies with several products may apply to multiple categories, though the organisers advise concentrating on the strongest product in the most relevant category.
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: The programme is open to startups rather than academic groups or individuals, and the published criteria are a clear physical AI use case, a minimum viable product in active use or testing, a registered legal entity, and a live public website. The MVP requirement is the real filter: this is not a competition for a research prototype or a paper, and a team whose work exists only as published results with no product in the hands of users or testers does not meet the bar regardless of technical quality. The emphasis throughout is on demonstrable product, meaningful traction, and a credible plan to scale, which means the submission should foreground deployment evidence, users and a scaling argument rather than benchmark performance. No geographic restriction is stated, so the competition appears open globally, which distinguishes it from the national innovation agency funding that otherwise dominates robotics and embodied AI support. Each application covers a single product, and a company with several distinct products may submit to more than one of the five categories, though the organisers explicitly recommend leading with the strongest product in the single most relevant category rather than entering broadly. Applications close on 25 October 2026 and finalists and winners are announced in mid-November 2026. Applicants should verify current terms and the application link directly with Nebius, since the programme is newly launched and detailed terms and conditions may be updated during the open window. Confirm the full requirements in the official notice before applying.
The current listing shows each of the five category winners receives 150,000 US dollars in Nebius compute credits, so amount_min and amount_max are both recorded as 150,000 US dollars, with a total prize pool of 750,000 dollars across the five categories. The award is compute rather than cash, and for a physical AI company that distinction cuts both ways. On the favourable side, 150,000 dollars of GPU time on current-generation hardware is worth considerably more to a robotics or world-model team than the same sum in cash would be after the overhead of procuring access, and it arrives without equity, without a term sheet and without the dilution that a comparable accelerator cheque would carry. On the constraining side, credits cannot pay salaries, buy robot hardware or fund field trials, which are precisely the costs that dominate a physical AI budget, so a team whose bottleneck is hardware or data collection rather than training throughput gains less than the headline figure suggests. The structure also means the award is tied to a single vendor's cloud, and a company already committed to another provider must weigh migration cost against the credit value. Companies may apply to more than one category if they have several distinct products, but each application covers one product, and the organisers advise leading with the strongest product in the single most relevant category rather than spreading applications, since a diluted submission competes against focused ones in every category it enters. Verify award ceilings, matching requirements, and allowable costs in the official notice.
Applications for Nebius and NVIDIA 2026 Physical AI Awards: Five $150,000 Compute Credit Prizes for Physical AI and Robotics Startups are due October 25, 2026. Build your timeline backwards from this date to cover registrations, approvals, and final submission checks.
Nebius and NVIDIA 2026 Physical AI Awards: Five $150,000 Compute Credit Prizes for Physical AI and Robotics Startups is funded by Nebius (NASDAQ: NBIS) in collaboration with NVIDIA. 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.
OpenAI's AI and Teen Development Research Grant Program makes up to 5,000,000 US dollars available in grants of up to 1,000,000 dollars for independent research on how AI affects adolescent development and wellbeing. Applications opened 8 September 2026 at 8:00 AM PDT and close 6 October 2026 at 11:59 PM PST, with notifications by 13 November 2026, and submission is through the SurveyMonkey Apply platform. The published scope has four strands: teen usage patterns and developmental outcomes; AI's potential effects on young people's lives; the factors that cause those effects to differ across populations and contexts; and design interventions and safety mechanisms. The inclusion of the fourth strand is what separates this from a pure effects-measurement programme, since it invites work that proposes and tests product-level safeguards rather than only documenting harm or benefit. The third strand is equally notable: OpenAI is asking specifically about heterogeneity of effect, which is an implicit acknowledgement that aggregate findings about teen AI use have limited policy value. Review is rolling and conducted by internal researchers and outside experts. Non-profit applicants are preferred and for-profit organisations are explicitly not prioritised, and indirect costs are capped at 10 percent of the grant, both of which shape who can realistically compete.
The Citi Foundation's 2026 Global Innovation Challenge commits 25,000,000 US dollars in grants of 500,000 dollars each to up to 50 nonprofit organisations, and AI is core to the challenge rather than one theme among many: the call is framed around economic opportunity in an economy increasingly shaped by AI. Three focus areas are published. The first is cultivating AI skills alongside human-centric competencies, a pairing that signals the Foundation is not funding tool training in isolation. The second is expanding digital access through community hubs offering shared devices and open-source AI tools, which is a deliberate bet on shared rather than individual access infrastructure. The third is building NGO capacity by integrating AI into youth employability programmes, meaning the grantee's own operating model is in scope, not only its services. Geographic reach is approximately 70 countries where Citi operates across Africa, Asia Pacific, Europe, the Middle East, Latin America, the Caribbean and North America, but US applicants are restricted to 12 states plus the District of Columbia and Puerto Rico, which makes this substantially more open to international applicants than to most US ones. The process begins with a letter of inquiry due 6 October 2026 at 12:00 pm ET, one per organisation and in English only; organisations advancing to Round 2 must supply externally audited financials from 2024 or later.
The Atlassian Foundation's Collective Action open call commits 8,000,000 US dollars to fund partnerships of up to 1,000,000 dollars each over two years, and the 2026 round is framed explicitly around how the education sector should respond to AI. The Foundation is seeking initiatives that help education systems respond to AI so that the technology closes gaps for disadvantaged learners rather than widening them, and it states that a deliberate decision not to use AI can itself be a fundable response. That framing is unusual and worth taking at face value: the call is not an AI-adoption fund, and a proposal whose theory of change is that more AI in classrooms will help disadvantaged learners is not what is being asked for. What is being asked for is collective action, meaning multiple players sharing goals and a learning agenda, directed at direct relief of disadvantage for an identified beneficiary group, in the context of an education and workforce-development sector being reshaped by AI. The lead applicant must be a registered nonprofit and may be based anywhere in the world, while other collective partners may come from any sector including government and industry, which makes this one of the few genuinely global pooled funds in the AI-and-education space. Applications close 30 October 2026 at 11:59 pm AEDT and are submitted through the Foundation's Goodstack nomination portal; applicants must pass an eligibility checklist on which any single negative answer is disqualifying.
NVIDIA committed $250 million over five years to UK agentic safety and security research, delivered as compute access through UK-based Cloud Partners. No open call, no cash, and the commercial terms are not yet agreed. A breakdown of compute-as-grant and how to budget for it.
Read articleNSF 26-513 makes roughly $100 million available for up to 10 State and Regional AI Infrastructure Hubs at $4M to $12M each over five years. One award per state or multi-state region. One proposal per organization. And NSF is not buying you GPUs — it funds the coordination, the workforce and the faculty training, while the compute has to come from partners you have to already have.
Read articleWith Forms J in FY2027, NIH will no longer accept letters of support — the boilerplate endorsements applicants have padded proposals with for decades. Notice NOT-OD-26-094 replaces them with letters of collaboration, aligning NIH with NSF. Here is exactly what changes, what a compliant letter must now contain, who can no longer write one, and how this fits the broader re-engineering of the NIH application package.
Read article