Sectoral AI Governance Act of 2026
HB9125, titled the Sectoral AI Governance Act of 2026, would give the heads of federal agencies explicit authority to issue rules governing the use of algorithmic decision-making systems when the agency determines, based on available evidence, that those systems are likely to materially contribute to violations of federal law the agency already enforces. The bill is structured as a sector-by-sector regulatory framework rather than a single cross-government AI regime, and it applies to systems used to make or assist decisions, including those based on machine learning, statistics, or artificial intelligence, while excluding passive computing infrastructure.
The bill requires agencies to use notice-and-comment rulemaking, generally preceded by an advanced notice of proposed rulemaking, and to consult with OIRA, OSTP, and NIST as appropriate. It also directs agencies to coordinate to avoid conflicting requirements, consider impacts on government services and public benefits, periodically review and update rules, and report to Congress and the public on rulemakings, enforcement actions, staffing needs, technical assessments, and any disparate impacts or discriminatory effects identified. A violation of a rule issued under the bill would be treated as a violation of the underlying federal law for administrative and civil enforcement purposes.
The bill would expand and clarify agency rulemaking authority under existing federal enforcement statutes by authorizing prospective rules targeted at algorithmic systems that materially contribute to legal violations. It would not create a single new AI regulator, but instead would affect any federal agency with enforcement authority, potentially influencing sectors such as consumer protection, employment, housing, lending, civil rights, health, transportation, and other regulated areas. The bill also preserves state authority to regulate algorithmic decision-making systems except where state law conflicts with the Act or a rule issued under it.
Based on the bill text and the absence of recorded committee debate or votes, the overall posture appears policy-driven and preventive rather than overtly partisan in the available record. The findings emphasize regulatory clarity, coordination, transparency, and accountability, suggesting the bill is framed as a governance and enforcement tool to address AI-related harms. No vote history or transcript evidence is available to show support or opposition levels, but the structure of the bill indicates an attempt to balance stronger federal oversight with procedural safeguards and interagency coordination.
The main points of contention are likely to be the breadth of agency authority, the standard for determining when an algorithmic system is likely to materially contribute to a violation, and whether the bill could lead to overlapping or inconsistent regulations across agencies. Another likely issue is the potential compliance burden on regulated entities and the administrative burden on agencies, especially given the reporting, consultation, and periodic review requirements. At the same time, supporters would likely emphasize the bill’s safeguards—public notice, consultation with technical experts, coordination across agencies, and explicit protection for state authority absent conflict—as ways to reduce arbitrary or duplicative regulation.