HR-9381-119
Ordered to be Reported (Amended) by the Yeas and Nays: 19 - 14.
Sponsored by Tim Walberg (R-MI)
What it does
This bill would amend the 1913 law establishing the Department of Labor to require the Bureau of Labor Statistics (BLS) to collect and report data on the use of artificial intelligence in the workplace. It would direct BLS to measure AI adoption by employers, AI use by workers within their jobs, AI-related changes to tasks and work organization, AI-related training, and AI's effects on hiring, wages, and employment levels. BLS would be required to begin collecting this data within 18 months of enactment.
Who benefits
Workers seeking to understand how AI may affect their jobs or wages. Employers and businesses that use labor market data for workforce planning. Educators and workforce training providers who need data to design relevant programs. Policymakers and researchers studying AI's economic effects. Job seekers navigating a changing labor market. Economic forecasters and employment projection services that rely on BLS data. Labor unions monitoring AI's effects on their members.
Who is hurt
Employers who may face new survey burdens from additional BLS data collection requirements. BLS itself, which would need to design, fund, and staff new data collection programs within a fixed 18-month window — potentially straining existing resources. Taxpayers who may bear the cost of expanded federal statistical operations, though the bill does not include an explicit appropriation. Workers in industries where newly published AI adoption data could accelerate automation investment by highlighting high-automation-potential roles.
Supporters argue
Supporters argue that the federal government currently lacks timely, consistent data on how AI is reshaping employment, wages, and job tasks — a gap that leaves policymakers, educators, and workers flying blind during a period of rapid technological change. They contend that BLS is the appropriate agency for this mission, given its existing infrastructure for employer- and worker-side surveys, and that better data would allow workforce training programs, schools, and businesses to respond more effectively to labor market shifts before displacement becomes widespread.
Opponents argue
Opponents argue that the bill mandates a new, open-ended data collection program without authorizing any funding, leaving BLS to absorb the costs from an already constrained budget — potentially degrading the quality of existing statistical programs that millions of users depend on. They contend that "artificial intelligence" is a rapidly evolving and contested concept, and that locking a statutory definition into the 1913 Department of Labor Act may produce data that quickly becomes outdated or methodologically inconsistent as the technology changes.
Constitutional context
No significant constitutional tension exists here. Congress has clear authority under the Commerce Clause (Art. I, §8, cl. 3) and the Necessary and Proper Clause (Art. I, §8, cl. 18) to direct a federal agency to collect labor market statistics. This bill does not delegate broad rulemaking authority to an agency, so the major questions doctrine from West Virginia v. EPA (2022) and the post-Chevron scrutiny from Loper Bright v. Raimondo (2024) are not directly implicated.
Checks and balances
Congress gains a statutory mandate over BLS data collection priorities; BLS (executive branch) retains discretion over methodology and implementation, subject to congressional oversight and appropriations.
Historical precedent
Congress has previously directed BLS to expand data collection in response to emerging labor market trends, including the creation of the Contingent and Alternative Employment Arrangements supplement in the 1990s to measure the rise of non-traditional work arrangements.