United StatesArtificial Intelligence Risk Management and Security Act of 2026
An AI governance rule binding private bodies.
- Obligation class
- Governance, Reporting, Disclosure
As of .
What it requires
- Develop, publish and follow a Model Safety Plan for each AI system or model you create, substantially modify, or use in the training or evaluation of other AI models, naming the model or system, a capabilities or risk assessment, the mitigation measures for each assessed item across the product lifecycle, and the corporate officer responsible for carrying it outNIST AI RMFMANAGE 1.3NIST AI RMFGOVERN 2.3
- Submit each Model Safety Plan to the Secretary of Commerce in the form and manner the Secretary determinesNIST AI RMFGOVERN 1.1
- Comply with each standard the Secretary of Commerce adopts by rule that applies to youNIST AI RMFGOVERN 1.1
- Make a frontier AI model, including its weights, configuration files, runtimes and software libraries needed to operate it, available to the Artificial Intelligence Safety Board at least 45 calendar days before you introduce it into interstate or foreign commerceNIST AI RMFGOVERN 1.1
- If you are a developer or provider of a frontier AI model, or an operator of critical infrastructure that uses an AI model to manage industrial control systems or other operational technologies, report any confirmed AI safety incident or AI security incident within 30 days of confirmation, and within 72 hours if the incident poses an imminent threat to national security, critical infrastructure or public safety.NIST AI RMFMANAGE 4.3
These lines are LexLint's reading of the law's text, as of .
In the frameworks
- NIST AI RMF
- GOVERN 1.1Legal and regulatory requirements involving AI are understood, managed, and documented.
- GOVERN 2.3Executive leadership of the organization takes responsibility for decisions about risks associated with AI system development and deployment.
- MANAGE 1.3Responses to the AI risks deemed high priority, as identified by the MAP function, are developed, planned, and documented. Risk response options can include mitigating, transferring, avoiding, or accepting.
- MANAGE 4.3Incidents and errors are communicated to relevant AI actors, including affected communities. Processes for tracking, responding to, and recovering from incidents and errors are followed and documented.
- NIST AI 600-1
- GAI-RISK-07Human-AI Configuration
- GAI-RISK-09Information Security
- NIST AI 200-2 Draft
- OWASP LLM Top 10
- LLM03:2026Excessive Agency
- OWASP Agentic Top 10
- ASI09Human-Agent Trust Exploitation
- MIT mitigations Draft
- 1.1Board Structure & Oversight
- 1.2Risk Management
- 3.5Post-deployment Monitoring
- 4.1System Documentation
- 4.2Risk Disclosure
- 4.3Incident Reporting
1 more
- 4.5Third-Party System Access
- NIST CSF 2.0
- RS.CO-02Internal and external stakeholders are notified of incidents
- FINOS AIGF
- AIR-PREV-005System Acceptance Testing
- AIR-PREV-006Data Quality & Classification/Sensitivity
- AIR-PREV-024Human-in-the-Loop Action Approval Gate
- AIR-DET-004AI System Observability
- AIR-DET-011Human Feedback Loop for AI Systems
- AIR-DET-021Agent Decision Audit and Explainability
- ETSI EN 304 223
- 5.1.4-1Capability for human oversight designed in and maintained.
- 5.2.1-1Full inventory of assets, including their dependencies and connections.
- 5.2.2-5AI-specific incident management and recovery plans that are created, tested and maintained.
- 5.2.4-1Developers keep an audit trail of system design and post-deployment maintenance plans and offer it downstream.
- 5.4.2-1Operators log system and user actions.
Our reading of where this law's duties sit in each framework, and in the NIST AI RMF, line by line beside each duty above. It never says that running a control meets the law. How we read a framework against the law
Identifiers refer to ETSI EN 304 223 V2.1.1 (2025-12), © ETSI 2025, all rights reserved. ETSI owns the standard; no ETSI text (titles or sentences) is reproduced here. The standard is free to download from ETSI.
If you get it wrong
Penalty structure
Section 3(d)(2) caps the civil penalty at $250,000 for each violation of the duty to comply with an adopted standard, and section 3(d)(3) makes each day a violation continues a separate violation.
- Rule
- Per violation only
- As of
- Currency
- USD
- Per violation unit
- Violation
- Per violation amount
- 250,000
Who enforces it
Enforcement body
Secretary of Commerce, who enforces compliance with the standards; the Attorney General may sue to enjoin a violation or recover the civil penalty at the Secretary's request.
What this law does
The bill was introduced on by Senators Warner, Schatz and Kim and referred to the Senate Committee on Commerce, Science, and Transportation. It would establish an Artificial Intelligence Safety Board within the Department of Commerce. It would require each developer of a frontier artificial intelligence model to develop, publish and follow a safety plan for each AI system or model it creates, substantially modifies or uses in training or evaluating other models.
It would also require each developer to comply with the standards the Secretary of Commerce adopts, subject to a civil penalty of up to $250,000 for each violation. A developer would have to give the Board access to a frontier model at least 45 calendar days before introducing it into interstate or foreign commerce.
When LexLint raises it
When your app profile says your app trains models.