# NIST SP 800-63-4 digital identity guidelines, explained

The fourth revision of NIST’s Digital Identity Guidelines, final in July 2025, keeps three assurance levels each for identity proofing, authentication and federation but rebuilds the risk process around them, and OMB M-26-18 now requires federal agencies to run that process.

Explainer · Cross-sector · Customer & Citizen Experience. Published November 14, 2026. Last reviewed November 14, 2026.

## At a glance

- NIST SP 800-63-4, Digital Identity Guidelines, published in final form on July 31, 2025 (NIST’s document history). It supersedes SP 800-63-3.
- A suite: the base volume (risk management and assurance levels) plus SP 800-63A-4 (identity proofing), 800-63B-4 (authentication) and 800-63C-4 (federation).
- Three dimensions of assurance: Identity Assurance Level (IAL), Authentication Assurance Level (AAL) and Federation Assurance Level (FAL), each at levels 1 to 3.
- Its change log lists an expanded digital identity risk management (DIRM) process that allows tailoring, performance metrics, redress, AI and machine learning, and a user-controlled wallet federation model. Each online service needs a Digital Identity Acceptance Statement.
- OMB M-26-18 (August 31, 2026) requires agencies to conduct a DIRM process consistent with SP 800-63-4 for public-facing services within 240 days, and to offer Login.gov as a sign-on option on in-scope public-facing websites.

## What changed from SP 800-63-3

SP 800-63-3 introduced separate assurance levels for identity proofing, authentication and federation, and split the guidance into a suite of four documents [1]. SP 800-63-4 keeps that structure. Its change log says it substantially updates and reorganizes SP 800-63-3, including [1]:

- expanded security, privacy and customer experience considerations;
- updated digital identity models, adding a user-controlled wallet federation model to address digital wallets and attribute bundles;
- an expanded digital identity risk management process that defines protected online services, user groups and impacted entities, and allows additional assessments to tailor initial baseline controls;
- performance metrics for the continuous evaluation of digital identity systems;
- a new subsection on redress processes and requirements; and
- a new subsection on the use of AI and machine learning in digital identity services.

NIST’s document history lists drafts dated 06/08/20, 12/16/22 and 08/21/24, and the final on 07/31/25. A planning note dated 10/06/2026 says an Implementation Resource Hub for SP 800-63-4 is now available [2].

## Who the guidelines are for

The guidelines apply to all online services for which some level of assurance in a digital identity is required, whatever the constituency: the public, business partners, or government employees and contractors. They focus mainly on services for external users, such as people applying for public benefits, but also apply to federal systems used by employees and contractors. National security systems are not covered [1]. Private-sector organizations and state, local and tribal governments may consider using them where appropriate [1]. The guidelines do not explicitly address machine-to-machine authentication, Internet of Things devices or API access on behalf of subjects [1].

## IAL, AAL and FAL

The base volume summarizes each level; the normative requirements are in the companion volumes [1].

- **Identity assurance.** IAL1 supports the real-world existence of the claimed identity and provides some assurance that the applicant is associated with it, with core attributes validated against authoritative or credible sources. IAL2 adds evidence and more rigorous validation and verification. IAL3 adds an on-site attended session with a trained proofing agent and the collection of at least one biometric [1].
- **Authentication assurance.** AAL1 allows single-factor or multi-factor authentication, and verifiers are expected to offer and encourage multi-factor options. AAL2 requires two distinct factors and approved cryptography, and the summary table lists “offer phishing-resistant options” among its control objectives. AAL3 requires a public-key cryptographic authenticator with a non-exportable private key that provides phishing resistance [1].
- **Federation assurance.** FAL1 protects against forged assertions. FAL2 adds protection against injection attacks. The stated objective for FAL3 is protection against compromise of the identity provider [1].

Each summary table pairs a level with a user profile. For example, IAL2 fits services where users can view and change financial information such as a direct deposit location; AAL3 fits services where privileged access could result in a breach that would be a major incident under OMB guidance [1].

## The risk process and the acceptance statement

The guidelines say they promote a risk-based approach to implementation rather than a compliance-oriented one, and encourage organizations to tailor controls [1]. The DIRM process runs through five steps: define the online service, conduct an initial impact assessment, select initial assurance levels and baseline controls, tailor and document the levels, and continuously evaluate and improve. Organizations must develop and document a process and governance model for selecting initial assurance levels and controls [1].

The output is a **Digital Identity Acceptance Statement**. Organizations must prepare one for each online service they manage and for each external online service that supports their mission, including software-as-a-service offerings. At a minimum it records the initial impact assessment, the initially assessed assurance levels, any tailored level with its rationale, compensating controls with their comparability or residual risk, and supplemental controls. Federal agencies should include it in the system authorization package [1]. A relying party that uses an external credential service provider must review that provider’s statement and fold the relevant parts into its own [1].

In our view, the acceptance statement is the most useful change for practitioners, because it records why a service accepted a given level of assurance and what it added to compensate. That makes a later review, or a fraud investigation, much easier to run.

## Federal policy that points to it: OMB M-26-18

On August 31, 2026, OMB issued M-26-18, Scaling Use of Login.gov to Deliver a Universal Sign-on for Public Services [3]. It covers public-facing agency websites where the public signs in, and not sites used solely by organizations or by people acting for others. Its main requirements [3]:

- agencies must offer Login.gov as a sign-on option for all in-scope public-facing websites, as required by 6 U.S.C. 1523(b)(1)(D), or, if they cannot meet the deadlines, send OMB a notice containing a personal certification from the agency head;
- where identity verification is also needed, agencies must use Login.gov’s verification service unless it does not meet the needs of the service, and may offer other solutions only in defined cases, seeking parity in assurance levels;
- in choosing whether to require verification and at what level, agencies must consider the effect on users, avoid friction the risk does not justify, document their risk-based decisions and monitor outcomes over time; and
- agencies using Login.gov for verification should accept its suitable offerings at the IAL the agency sets for the service.

The memo refers to SP 800-63-4 throughout: for its glossary, for the DIRM method of selecting assurance levels, for sample performance metrics, and to the companion volumes for the levels themselves [3]. Its appendix sets agency deadlines: an inventory of public-facing websites with authentication within 60 days; a DIRM process consistent with SP 800-63-4 for public-facing services within 240 days; Login.gov on in-scope High Impact Service Provider websites within one year; and Login.gov on all existing in-scope websites within two years, with a notice to OMB as the alternative for the last two. NIST is to publish a DIRM resource within 120 days [3].

The memo also says agencies should follow existing OMB policy to manage digital identity, naming M-19-17, M-21-04 and M-22-09, and it supersedes section V.5 of M-19-17 on shared identity services [3]. It states that the statutory Login.gov mandate and certification requirement, under 6 U.S.C. 1523(c), do not apply to the Department of War, national security systems or elements of the intelligence community [3].

## Status

As checked on November 14, 2026, the latest entry in the document history on NIST’s publication page is the final SP 800-63-4, [2], and the M-26-18 PDF is available on whitehouse.gov [3]. This explainer describes the guidelines and the memo; it is not legal advice, and organizations should confirm how they apply with counsel.

Moraph’s teams help organizations with work like this. [See our solutions](https://www.moraph.com/solutions).

## Sources

1. National Institute of Standards and Technology. NIST SP 800-63-4, Digital Identity Guidelines (July 2025). <https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-63-4.pdf> Checked November 14, 2026.
2. NIST Computer Security Resource Center. SP 800-63-4, Digital Identity Guidelines (publication page and document history). <https://csrc.nist.gov/pubs/sp/800/63/4/final> Checked November 14, 2026.
3. Office of Management and Budget. M-26-18, Scaling Use of Login.gov to Deliver a Universal Sign-on for Public Services (August 31, 2026). <https://www.whitehouse.gov/wp-content/uploads/2026/08/M-26-18-Scaling-Use-of-Login.gov-to-Deliver-a-Universal-Sign-on-for-Public-Services.pdf> Checked November 14, 2026.

Published: November 14, 2026
Canonical URL: <https://insights.moraph.com/nist-sp-800-63-4-digital-identity-guidelines-explained>
Publisher: Moraph Insights
