Congress quietly builds AI safety database—but keeps reporting voluntary

What it does

This bill directs the National Institute of Standards and Technology (NIST) to establish a voluntary program for AI companies and researchers to report flaws, vulnerabilities, and safety issues in artificial intelligence systems. NIST will convene industry, academia, and government to develop common definitions, technical standards, detection methods, and a national database to track AI flaws—benefiting the public by improving AI safety and security transparency without imposing mandatory reporting requirements.

Why we flagged it

The bill's core mechanism is establishing voluntary reporting infrastructure and common standards for AI flaws—a public-interest transparency and safety measure, not a regulatory mandate or industry carve-out.

What the text implies

• Voluntary reporting may create a de facto industry standard that smaller AI developers struggle to meet, potentially concentrating market power among larger firms with compliance resources.
• The national database and standardized taxonomies could become a regulatory baseline for future mandatory disclosure rules, shifting from voluntary to compulsory over time.
• NIST's authority to designate 'eligible entities' to maintain the database may route data collection through private contractors or academic institutions, creating intermediaries between companies and government oversight.
• The bill does not address liability protection for companies reporting flaws, leaving open the question of whether disclosed vulnerabilities could be used in litigation—potentially chilling voluntary participation.

Section numbers refer to the bill text the analysis read — linked under Primary records below.

Who it affects

The bill creates infrastructure for voluntary disclosure and standardization of AI safety and security issues, improving public visibility into AI risks and enabling faster detection and remediation. Citizens benefit from safer, more transparent AI systems, though the voluntary nature means participation depends on industry cooperation.

Who stands to gain

• AI software and hardware vendors (AMD, HPE, Cisco, Cognizant, Verisign)
• Cybersecurity and compliance consulting firms
• Academic institutions and research consortia designated as eligible entities
• Standards development organizations

Named in the bill

National Institute of Standards and Technology (NIST), Cybersecurity and Infrastructure Security Agency (CISA), Department of Homeland Security, AMD, HPE, Cisco, Cognizant, Verisign, Higher Education Act of 1965, National Artificial Intelligence Initiative Act of 2020, Advancing American AI Act

Where it stands

2 cosponsors: 1 Democrats, 1 Republicans.

• Jun 18, 2026 — Introduced · Congress.gov: “Introduced in House”
• Jun 18, 2026 — Referred to House Committee on Science, Space, and Technology · Congress.gov: “Referred to the House Committee on Science, Space, and Technology”
• Jun 25, 2026 — Markup held in committee · Congress.gov: “Committee Consideration and Mark-up Session Held”
• Jun 25, 2026 — Reported out of committee · Congress.gov: “Ordered to be Reported in the Nature of a Substitute by the Yeas and Nays: 35 - 0”

Dates and quoted wording are Congress.gov's action record; the timeline shows status changes, not every procedural step.