National AI Research Resource (NAIRR)

The National AI Research Resource (NAIRR) is a US federal initiative -- currently operating as the NAIRR Pilot -- that provides researchers, educators, and students with shared access to AI computational infrastructure (supercomputing and cloud allocations), AI-ready datasets, pretrained models, and educational resources, rather than direct cash funding. It is led by the National Science Foundation (NSF) in partnership with more than a dozen other federal agencies, including the Department of Energy (DOE) and National Institutes of Health (NIH), who co-lead a security-focused 'NAIRR Secure' track for sensitive data. Eligible US-based researchers, educators, and graduate students (with a faculty sponsor) apply for time-limited resource allocations -- typically a 3-month Start-Up award or a 12-month Research award -- through a competitive, proposal-based process rather than an open-access account. A project 'uses NAIRR' when it draws on the pilot's pooled portfolio of partner-provided compute, storage, or model resources under an awarded allocation; a project that purchases or funds its own AI hardware, or that uses a single university's internal HPC cluster with no NAIRR allocation, is not a NAIRR activity even if the research itself is AI-related.

ByCASRAI Editorial Board

· Last updated 18 Jul 2026

Worked examples

Is an instance

A university AI researcher without institutional access to large-scale GPU infrastructure submits a Start-Up Project Request through nairrpilot.org and is awarded a 3-month allocation on a partner supercomputing platform (such as Delta, Bridges-2, or Jetstream2) to train and evaluate a model.

Is an instance

A graduate student at a non-R1 institution, with a support letter from a faculty advisor, requests access to curated AI-ready datasets and pretrained models through the NAIRR Pilot's resource catalog for a course project or thesis research, rather than purchasing commercial cloud compute.

Looks similar, but isn't

Not an instance

A lab that uses NSF grant funds to purchase its own on-premise GPU cluster is not using NAIRR -- NAIRR provides shared access to an existing, partner-contributed resource portfolio, not direct funding for equipment purchases.

Not an instance

A researcher running AI workloads on a university's internally funded HPC cluster, with no NAIRR-awarded allocation involved, is not engaged in a NAIRR activity, even though the work is AI-related.

The National AI Research Resource (NAIRR) is a US federal initiative — currently operating as the NAIRR Pilot — designed to give researchers and educators outside a handful of well-resourced institutions shared access to the computational infrastructure, data, and models needed for AI research. Rather than a grant program that disburses funds, NAIRR is an access program: successful applicants receive a time-limited allocation against a pooled portfolio of compute, storage, and model resources contributed by federal agencies, national labs, and industry/nonprofit partners.

Origin and leadership

NAIRR traces to a National AI Research Resource Task Force established by Congress in 2021, which delivered its final implementation roadmap in January 2023. The NAIRR Pilot launched in January 2024 under the leadership of the National Science Foundation (NSF), in partnership with more than a dozen other federal agencies — including the Department of Energy (DOE), the National Institutes of Health (NIH), NASA, NIST, and others — several of which contribute compute, data, or domain expertise to the shared resource pool. DOE and NIH co-lead a separate, more restricted NAIRR Secure track intended for AI research involving sensitive, controlled, or otherwise non-public data.

What resources are available

Through the pilot’s portal at nairrpilot.org, awarded projects can draw on: national supercomputing and cloud allocations (partner systems have included Delta, Bridges-2, Frontera, Summit-class resources, the Argonne AI Testbed, and Jetstream2); curated AI-ready datasets and pretrained models; and educational/training materials aimed at classroom and early-career use. This is distinct from a Data Management Plan (DMP), which governs how a funded project will manage its own data — a NAIRR allocation is itself a resource a DMP might describe using.

Who is eligible and how allocations work

The pilot is open to US-based researchers and educators at academic institutions, non-profits, federal agencies or FFRDCs, state/local/tribal agencies, and startups or small businesses with federal grants; graduate students may apply with a support letter from a faculty advisor. Two main allocation tracks exist: a Start-Up Project (roughly 3 months, faster turnaround, intended to let a team demonstrate feasibility) and a full Research Project (roughly 12 months). Proposals are reviewed on a rolling monthly cycle rather than a single annual deadline. Because allocations are competitively awarded and resource-specific, NAIRR access is not equivalent to a standing institutional compute account — a team must reapply or request renewal as its allocation period ends.

Pilot status and the push for permanent authorization

As of mid-2026, NAIRR remains formally a pilot rather than a permanently authorized program. NSF reports the pilot has supported several hundred research projects and thousands of students across all 50 states, Washington DC, and Puerto Rico, and has released a solicitation to establish a NAIRR Operations Center intended to carry the effort from pilot into a more sustained operating model. Separately, bipartisan legislation — the CREATE AI Act — has been reintroduced in Congress to place NAIRR on a permanent statutory footing within NSF, to be operated by a competitively selected nongovernmental entity; as of this writing the bill has not passed and its status should be treated as developing rather than settled (REPORTED, not confirmed enacted). Research administrators tracking NAIRR for compliance or planning purposes should confirm current program status directly at nairrpilot.org and nsf.gov rather than assuming pilot terms are final.

NAIRR in context

NAIRR sits alongside, rather than replaces, other routes to AI research infrastructure: institutional research security and export-control review still apply to NAIRR-enabled work involving sensitive technology or foreign collaboration, and a NAIRR allocation does not substitute for the funding a grant provides — it supplies the compute and data access a funded project may need to carry out AI-related aims. Research offices should treat a NAIRR award as a resource commitment to be documented and cited in the same way as any other externally provided infrastructure, and should note it separately from cash awards when reporting project support.

Use in your systems

JATS XML <role> element

xml

<role vocab="credit"
      vocab-identifier="https://casrai.org/dictionary/"
      vocab-term="National AI Research Resource (NAIRR)"
      vocab-term-identifier="https://casrai.org/dictionary/term/national-ai-research-resource-nairr" />

Schema.org DefinedTerm (JSON-LD)

json

{
  "@context": "https://schema.org",
  "@type": "DefinedTerm",
  "@id": "https://casrai.org/dictionary/term/national-ai-research-resource-nairr",
  "name": "National AI Research Resource (NAIRR)",
  "identifier": "https://casrai.org/dictionary/term/national-ai-research-resource-nairr",
  "description": "The National AI Research Resource (NAIRR) is a US federal initiative -- currently operating as the NAIRR Pilot -- that provides researchers, educators, and students with shared access to AI computational infrastructure (supercomputing and cloud allocations), AI-ready datasets, pretrained models, and educational resources, rather than direct cash funding. It is led by the National Science Foundation (NSF) in partnership with more than a dozen other federal agencies, including the Department of Energy (DOE) and National Institutes of Health (NIH), who co-lead a security-focused 'NAIRR Secure' track for sensitive data. Eligible US-based researchers, educators, and graduate students (with a faculty sponsor) apply for time-limited resource allocations -- typically a 3-month Start-Up award or a 12-month Research award -- through a competitive, proposal-based process rather than an open-access account. A project 'uses NAIRR' when it draws on the pilot's pooled portfolio of partner-provided compute, storage, or model resources under an awarded allocation; a project that purchases or funds its own AI hardware, or that uses a single university's internal HPC cluster with no NAIRR allocation, is not a NAIRR activity even if the research itself is AI-related.",
  "inDefinedTermSet": "https://casrai.org/dictionary/domain/ai-ml-research-outputs#set",
  "url": "https://casrai.org/dictionary/term/national-ai-research-resource-nairr",
  "sameAs": [],
  "license": "https://creativecommons.org/licenses/by/4.0/",
  "publisher": {
    "@id": "https://casrai.org/#organization"
  },
  "dateModified": "2026-07-18T11:05:51",
  "inLanguage": "en"
}