AI Consortium FAQs

The National Institute of Standards and Technology (NIST) previously announced the formation of the AI Safety Institute Consortium (AISIC) through a publication in the Federal Register dated November 2, 2023 (88 FR 75276). As part of NIST’s efforts to broaden the scope of the Consortium’s collaborative research undertakings, the retitled NIST Artificial Intelligence Consortium’s new name reflects the group’s augmented goals. While much of the Consortium’s previous work will continue, NIST hopes to both refine and expand the Consortium’s concentration on AI metrology, standards development, technological innovation, investment, security, and economically beneficial adoption.

Participation in the Consortium is open to all interested organizations that can contribute their expertise, products, data, and/or models to the activities of the Consortium.

Selected participants will be required to enter into a Consortium Cooperative Research and Development Agreement (CRADA) with NIST. At NIST’s discretion, entities that are not legally permitted to enter into CRADAs may be allowed to participate in the Consortium under a separate non-CRADA agreement.

All applicants for participation in the Consortium will be screened and verified in accordance with NIST’s internal processes and procedures prior to being invited to sign a CRADA (or other agreement) with NIST and become a Consortium Member.

Yes, organizations outside of the United States are encouraged to submit letters of interest in accordance with the instructions published in the Federal Register on May 29, 2026 (91 FR 32007). All applicants for participation in the Consortium will be screened and verified in accordance with NIST’s internal processes and procedures, and applicants will be contacted by NIST when such vetting is complete.

No, individuals are not eligible to join the Consortium in their personal capacities. The Consortium is open only to organizations that are legal entities.

Interested parties should submit a letter of interest to NIST in accordance with the instructions published in the Federal Register on May 29, 2026 (91 FR 32007). All applicants for participation in the Consortium will be screened and verified in accordance with NIST’s internal processes and procedures, and applicants will be contacted by NIST when such vetting is complete.  

No, there is no deadline for submitting a letter of interest for participation in the Consortium.

NIST will accept letters of interest submitted in accordance with the instructions published in the Federal Register on May 29, 2026 (91 FR 32007), on an ongoing basis, with regular periods of review for new applicants. The first period of review will begin within 60 days of the May 29^th Federal Register Notice, and selection determinations will subsequently be communicated to applicants.

No, all individuals participating in the Consortium through an organization must be named and identified in the organization’s CRADA with NIST. This applies to ALL individuals contributing to the efforts of the Consortium, even those participating on only an informal, temporary, or provisional basis, as well as any individuals who assist in the performance of the Consortium’s research without otherwise interacting with NIST and/or other Consortium Members.

To add or remove participating individuals from your organization’s CRADA with NIST, contact the NIST Consortium Manager to obtain an amendment to the CRADA.

The NIST AI Consortium will build on the work that started under the Consortium’s previous moniker – AISIC – in January 2024. With the announcement in the Federal Register of the Consortium’s retitling and expansion on May 29, 2026 (91 FR 32007), we anticipate that the Consortium’s activities will resume in earnest in August 2026. 

1.  FRN.

The NIST AI Consortium empowers the collaborative establishment of a new measurement science that will enable the identification of proven, scalable, and interoperable techniques and metrics to promote the development and use of AI. 

Although NIST anticipates many opportunities for the Consortium’s activities to broaden as the Consortium Members iteratively build upon the group’s work, NIST is currently preparing for research in six separate focus areas, detailed below.

1. The AI Testing, Evaluation, Verification and Validation (AI TEVV) Zero Draft Task Group will provide the community with tools to determine whether an AI system meets its design requirements and is sufficient for its intended use. “Zero drafts” are preliminary, stakeholder-driven drafts of standards that are as thorough as possible and will be submitted into the private sector-led standardization process.
2. The Annotation for AI Risks & Validity Task Group will support the development of a science-based tool kit for assessing the risks and impacts of AI for use with NIST’s Assessing Risks and Impacts of AI (ARIA) program.
3. The AI Evaluation and Measurement Methods Task Group will identify and prioritize gaps, barriers and open questions across the field of AI evaluation science based on input from a range of organization types, sectors and stakeholder roles.
4. The Bias Effects and Notable Generative AI Limitations (BENGAL) Group will team with IARPA to explore scalable solutions to misinformation, sensitive information leakage, flawed and inaccurate reasoning, and susceptibility to attack. The goal of BENGAL is to enable confident, effective use of large language models for intelligence analysis.
5. The AI Documentation Cards Task Group will provide standardized, practical templates for documenting AI datasets, AI models, AI systems and AI TEVV, based on existing best practices and community input.
6. NIST is also restarting the Chemical and Biological Security Task Group from the Consortium’s previous incarnation to share insights on emerging AI measurement and evaluation approaches relevant to chemical and biological security.

NIST will utilize the Consortium to empower the collaborative establishment of a new measurement science that will enable the identification of proven, scalable, and interoperable techniques and metrics to promote the development and use of AI. 

TPO has distributed draft CRADA amendments to all original Consortium Members with active CRADAs. If your organization has not yet received a draft CRADA amendment, please check all relevant inboxes for communications from agreements [at] nist.gov (agreements[at]nist[dot]gov). If your organization is unable to locate the relevant communication, inquiries should be addressed to agreements [at] nist.gov (agreements[at]nist[dot]gov) copying aiconsortium [at] nist.gov (aiconsortium[at]nist[dot]gov). 

Direct communications with requests for corrections to agreements [at] nist.gov (agreements[at]nist[dot]gov), copying aiconsortium [at] nist.gov (aiconsortium[at]nist[dot]gov) and identifying the point of contact at your organization to whom communications should be addressed. Following verification of all corrected/new information, NIST staff will issue a revised draft to the identified point of contact.

Direct communications with requests for updates to your organization’s name to agreements [at] nist.gov (agreements[at]nist[dot]gov), copying aiconsortium [at] nist.gov (aiconsortium[at]nist[dot]gov) and identifying the point of contact at your organization to whom communications should be addressed. Please include the organization’s new name (and DBA, if applicable), a brief explanation for the change, and specify whether the change is (a) a straightforward name change wherein the same DUNS/UEI have been retained, (b) the result of a merger/acquisition involving a new DUNS/UEI, or (c) some other scenario. Please share any relevant documentation (e.g., in the case of a merger/acquisition, include press release or official legal notice).

• Update to Principal Investigator (PI): Direct communications with requests to update your organization’s PI to agreements [at] nist.gov (agreements[at]nist[dot]gov), copying aiconsortium [at] nist.gov (aiconsortium[at]nist[dot]gov) and identifying the point of contact at your organization to whom communications should be addressed. Following verification of the new PI, NIST staff will issue a revised draft of this amendment to the identified point of contact. Do NOT execute a CRADA amendment that includes an incorrect or outdated PI name.

• Update to Project Team Personnel: Direct communications with requests to update your organization’s Project Team personnel to agreements [at] nist.gov (agreements[at]nist[dot]gov), copying aiconsortium [at] nist.gov (aiconsortium[at]nist[dot]gov) and identifying the point of contact at your organization to whom communications should be addressed. Following verification of the changes to the Project Team, NIST staff will issue a separate NEW amendment to the identified point of contact. This will occur after the current round of amendments has been completed. Please note, however, that new Project Team members cannot participate in the Consortium until NIST has returned a fully executed amendment to your organization in which the new members are named.

Maybe; this depends upon an organization’s internal/legal structure. The individual signing the CRADA amendment on behalf of an organization must be a representative authorized to act on the organization’s behalf and legally bind the organization. If an organization’s PI holds this legal authority, they may also sign the CRADA amendment.

The amended CRADA limits what members can say publicly about the Consortium to the following: 

[Collaborator] is collaborating with the National Institute of Standards and Technology (NIST) in the NIST Artificial Intelligence Consortium to establish a new measurement science that will enable the identification of proven, scalable, and interoperable measurements and methodologies to promote innovative and secure development and use of AI. NIST does not evaluate commercial products under this Consortium and does not endorse any product or service used. Additional information on this Consortium can be found in the Federal Register at 88 FR 75276 and 91 FR 32007.

Any further public statements should not be made. NIST will take appropriate action if it becomes aware of Collaborators’ making statements that go beyond the language contained in the CRADA amendment.

Absolutely, but only if the original source is NIST. We actively encourage members to reshare and repost official announcements and updates made directly by NIST regarding its AI efforts.

Yes. Please forward the draft text of your external piece to aiconsortium [at] nist.gov (aiconsortium[at]nist[dot]gov). The membership engagement staff will coordinate directly with the PAO to review and clear the request.