FRAUD Act of 2025

• Authority: United States Congress
• Parent authority: Federal government
• Jurisdiction: United States
• Collections: U.S. federal laws
• Most recent activity: Proposed
• Most recent activity date: 2025-10-21

Risk subdomain coverage

Minimal coverage — Brief mention but no specific mitigations, recommendations or governance measures relating to the risk subdomain are described.

• Lack of capability or robustness
• Lack of transparency or interpretability

All other risk subdomains had no coverage in this document.

Risk Coverage Summary

The document has minimal coverage of AI risk domains, with primary focus on AI system robustness (7.3) through requirements to reduce false positives and update fraud detection models. There is minimal attention to transparency through reporting requirements (7.4). The document is primarily about implementing AI for fraud detection rather than comprehensively addressing AI risks.

Sector coverage

Good coverage — Comprehensive governance measures or mitigations with clear procedures relating to the sector are explicitly described.

• Health Care & Social Assistance
• Public Administration (excl. National Security)

All other sectors had no coverage in this document.

Sectors Summary

The document primarily governs AI use within two sectors: Health Care and Social Assistance (specifically Veterans Health Administration healthcare services) and Public Administration (specifically the Department of Veterans Affairs). The governance is focused on implementing AI-based fraud detection in VA healthcare claims processing.

Legislative Status

Hard Law

Legislative Status Summary

This is a federal statute enacted by the U.S. Congress that creates binding legal obligations on the Secretary of Veterans Affairs using mandatory language throughout, with specific implementation deadlines and reporting requirements to Congressional committees.

Actors

Proposer

• Entity Types: AI_Governance_Actor
• Summary: The document is a Congressional bill as indicated by its legislative format and the enactment clause specifying it is enacted by both chambers of Congress.

Target

• Entity Types: AI_Governance_Actor, AI_Deployer
• Summary: The Act directs the Secretary of Veterans Affairs to implement and deploy an AI-based information technology system for fraud detection in VHA programs. Healthcare entities and providers submitting claims are indirectly affected by this system.

Enforcer

• Entity Types: AI_Governance_Actor
• Summary: The Secretary is responsible for implementing and operating the system as mandated by law. Congressional committees enforce compliance through oversight and annual reporting requirements.

Monitor

• Entity Types: AI_Governance_Actor
• Summary: Congressional committees receive annual reports for six years on the effectiveness, savings, and planned enhancements of the fraud detection systems, providing ongoing monitoring and oversight.

AI Lifecycle Stages

Lifecycle stages covered by the documents, as defined in the OECD Framework for the Classification of AI Systems

• Plan and Design
• Collect and Process Data
• Build and Use Model
• Verify and Validate
• Deploy
• Operate and Monitor

AI Lifecycle Summary

The document covers multiple AI lifecycle stages with primary emphasis on operation and monitoring. It specifies design requirements through mandated system functions, addresses data processing of claims data, requires use of machine learning models, and extensively covers operational monitoring through continuous analysis and annual reporting requirements.

Technical Scope

Terms relating to technical scope that are mentioned or implicitly referenced in the document

• AI Models
• AI Systems
• Predictive AI
• Task-specific AI

Technical Scope Summary

The document explicitly mentions AI systems (described as 'information technology system'), machine learning, and analytic models for predictive fraud detection. It focuses on task-specific predictive AI for fraud detection in healthcare claims. There is no mention of frontier AI, general purpose AI, foundation models, generative AI, compute thresholds, or open-source models.