Referred to the House Committee on Science, Space, and Technology.
The bill directs NIST, in consultation with CISA and multi‑stakeholders, to establish a voluntary program to define, report, track, and monitor "artificial intelligence flaws." It requires development of common definitions, taxonomies, technical standards, detection methods, reporting mechanisms, and infrastructure (including a national database or modification of an existing one). NIST must consult a broad set of stakeholders, may partner with eligible entities, consider interoperability and disclosure norms, and report to Congress within three years.
Voluntary program seen as insufficient (left) vs overreach risk (right).
Relative to its intended legislative type, this bill establishes a clearly scoped NIST-led administrative program to support voluntary reporting and monitoring of AI flaws, with specified activities, stakeholder engagement, and a required report to Congress.
The bill directs NIST, in consultation with CISA and multi‑stakeholders, to establish a voluntary program to define, report, track, and monitor "artificial intelligence flaws." It requires development of common definitions, taxonomies, technical standards, detection methods, reporting mechanisms, and infrastructure (including a national database or modification of an existing one).
NIST must consult a broad set of stakeholders, may partner with eligible entities, consider interoperability and disclosure norms, and report to Congress within three years.
The bill is voluntary and includes statutory definitions for AI, AI flaw, AI system, and eligible entities.
Technocratic, voluntary design improves prospects, but lack of funding language and procedural hurdles reduce near-term chances.
Relative to its intended legislative type, this bill establishes a clearly scoped NIST-led administrative program to support voluntary reporting and monitoring of AI flaws, with specified activities, stakeholder engagement, and a required report to Congress. It provides moderate procedural scaffolding but leaves key operational elements unspecified.
Voluntary program seen as insufficient (left) vs overreach risk (right).
These are examples from the analysis, not a ranked list of the most-affected groups.
Voluntary program seen as insufficient (left) vs overreach risk (right).
Generally supportive because it creates transparency, safety norms, and multi‑stakeholder oversight for AI risks.
Concerned that the bill is only voluntary and lacks strong enforcement, funding, or explicit equity and labor safeguards.
Cautiously positive: approves of multi‑stakeholder standards and voluntary approach but seeks clarity on funding, timelines, and security tradeoffs.
Wants measurable milestones and safeguards against misuse of reported data.
Skeptical: views federal creation of a national AI flaw database and expanded standards work as potential federal overreach and regulatory creep.
Prefers limiting federal role and protecting IP, trade secrets, and competitiveness.
Reached or meaningfully advanced
Reached or meaningfully advanced
Still ahead
Still ahead
Still ahead
Technocratic, voluntary design improves prospects, but lack of funding language and procedural hurdles reduce near-term chances.
The bill has not accumulated any surfaced votes yet.
Voluntary program seen as insufficient (left) vs overreach risk (right).
Technocratic, voluntary design improves prospects, but lack of funding language and procedural hurdles reduce near-term chances.
Relative to its intended legislative type, this bill establishes a clearly scoped NIST-led administrative program to support voluntary reporting and monitoring of AI flaws, with specified activities, stakeholder engagem…
Go beyond the headline summary with full stakeholder mapping, legislative design analysis, passage barriers, and lens-by-lens tradeoff breakdowns.