Title II — Measuring artificial intelligence capabilities, adoption and use, and layoff impacts

S. 3339 · 119th Congress · Dec 3, 2025 · Lineage

II Measuring artificial intelligence capabilities, adoption and use, and layoff impacts

Sec. 201 Support for evaluation of artificial intelligence automation

(a)

Prize competition required— Not later than 270 days after the date of the enactment of this Act, the Secretary of Commerce, acting through the Director of the National Institute of Standards and Technology (in this section referred to as the “Director”), shall commence carrying out at least 1 prize competition under section 24 of the Stevenson-Wydler Technology Innovation Act of 1980 (15 U.S.C. 3719) to develop benchmarks or similar reproducible methods to quantitatively measure the ability of artificial intelligence to automate or augment tasks or occupations, with the primary purpose of improving forecasts of the impacts that artificial intelligence may have on workers and the retraining needs of workers.

(b)

Consultation— In carrying out the prize competition required by subsection (a), the Director shall consult with the Secretary of Commerce, the Secretary of Labor, the Commissioner of the Bureau of Labor Statistics, and the Director of the National Science Foundation.

(c)

Structure and evaluation criteria—

(1)

In general— The Director shall develop the structure and evaluation criteria for a prize competition carried out under subsection (a) and a companion grant or cooperative agreement awarded under subsection (d) in accordance with the primary purpose described in subsection (a).

(2)

Competition scoping— The Director may structure a competition under subsection (a) into 1 or more categories, including by the occupation impacted or the capability domain of the artificial intelligence.

(3)

Category selection— In selecting the categories under paragraph (2), the Director shall consider—

(A)

the informativeness of a category in evaluating the direct or indirect impacts on labor markets, which may include factors such as the number of individuals directly or indirectly impacted by the automation or augmentation and the economic security or ability to find new employment opportunities of the individuals impacted;

(B)

the quality or efficacy of the submissions for a category;

(C)

the potential for a category to result in the development of benchmarks that complement existing benchmarks;

(D)

the potential for a category to result in the development of benchmarks that are underfunded by private contributions relative to the public value of the benchmarks; and

(E)

other factors determined appropriate by the Director, taking into account input gathered under section 101.

(4)

Competition phasing— The Director may structure a competition under subsection (a) into 1 or more phases (such as design and prototype stages).

(5)

Evaluation considerations— In developing the evaluation criteria for the benchmarks or similar methods submitted for prize competitions under subsection (a), the Director shall, to the extent practicable, seek to avoid or mitigate common problems affecting the value and feasibility of artificial intelligence benchmarks, including—

(A)

training data contamination that is difficult to detect;

(B)

rapid loss of discriminatory value over time due to a metric having a low ceiling for performance; and

(C)

other common problems, including problems identified in the input collected under section 101.

(d)

Companion grants and cooperative agreements— In carrying out this section, the Director may, in addition to carrying out a prize competition under subsection (a), award a grant or enter into a cooperative agreement to support the design, construction, validation, and maintenance of a benchmark or similar reproducible method described in subsection (a), including support for—

(1)

data collection and labeling;

(2)

evaluator training and rubric development; and

(3)

third-party replication and inter rater reliability testing.

(e)

Program administration— The Director may enter into contracts, cooperative agreements, or other agreements with for-profit or nonprofit entities, or State, territorial, local, or Tribal agencies, to design and administer prize competitions under subsection (a) and related grant and cooperative agreement activities under subsection (d).

(f)

Authorization of appropriations— There is authorized to be appropriated to the Secretary of Commerce to carry out this section $7,000,000 for the period of fiscal years 2026 through 2030.

Sec. 202 Voluntary artificial intelligence adoption and use reporting and data-sharing partnerships

(a)

In general— The Secretary, acting through the Commissioner of Labor Statistics and in coordination with the Secretary of Commerce and the Director of the Office of Science and Technology Policy, shall establish a program to enable and encourage artificial intelligence developers, artificial intelligence deployers, and other private entities, to share, on a voluntary basis, anonymized data on the adoption or use of artificial intelligence in the workforce.

(b)

Memoranda of understanding and other agreements— To carry out this section, the Secretary may enter into memoranda of understanding, data-licensing agreements, or other agreements for data sharing with artificial intelligence developers, artificial intelligence deployers, and other private entities, participating in the program under this section.

(c)

Use of data—

(1)

In general— Any data on adoption or use of artificial intelligence furnished through the program under this section—

(A)

shall be used exclusively for statistical purposes, as defined in section 3561 of title 44, United States Code;

(B)

shall be—

(i)

treated as confidential in accordance with the requirements under subchapter III of chapter 35 of title 44, United States Code; and

(ii)

exempt from disclosure under section 552(b)(3) of title 5, United States Code;

(C)

shall include an anonymized description of the data sources, the methodology used, and an assessment of the limitation of the data, including any known gaps in coverage or potential for bias; and

(D)

may not be used for regulatory enforcement or any antitrust action.

(2)

Statistical methodology— In establishing the program under subsection (a), the Secretary shall—

(A)

develop a statistical methodology to ensure that the data collected through such program is, to the extent feasible, statistically valid and representative of all sectors and demographics of the workforce;

(B)

establish data quality standards, uniform definitions, and formats for the data to be shared to ensure comparability; and

(C)

develop methods to identify and mitigate selection bias in the data received.

(d)

Public recognition for participation— The Secretary shall maintain, on the website of the Department of Labor, a public roster of artificial intelligence developers, artificial intelligence deployers, and other private entities, that participate in the program under this section and that agree to being included in such roster.

(e)

Publication of aggregate statistics—

(1)

In general— Not less frequently than once every 6 months, the Secretary shall publish machine readable statistics on information collected under this section, such as the degree of adoption or use of artificial intelligence for various occupations, unless the Secretary determines that such publications would be more detrimental than beneficial to the goal of collecting and sharing data to inform policymaking and the career decisions of individuals.

(2)

Aggregation— The data published under this subsection may be—

(A)

aggregated across data contributors; or

(B)

otherwise structured as the Secretary, in consultation with the Chief Statistician of the United States and the Workforce Information Advisory Council, determines appropriate.

(f)

Report to Congress— Not later than 2 years after the date of enactment of this Act, the Secretary shall submit to Congress a report—

(1)

evaluating—

(A)

participation rates of artificial intelligence developers, artificial intelligence deployers, and other private entities, in the program under this section;

(B)

the quality of data furnished through such program;

(C)

the usefulness of the aggregated statistics published pursuant to this section, including under subsection (e), for labor-market analysis; and

(D)

barriers to collection or analysis of data pursuant to this section; and

(2)

providing recommendations for expanding or modifying the program under this section.

(g)

Authorization of appropriations— There is authorized to be appropriated to carry out this section $7,000,000 for the period of fiscal years 2026 through 2030.

(h)

Definitions— In this section:

(1)

Artificial intelligence deployer— The term artificial intelligence deployer means an entity (including a cloud service provider) that operates artificial intelligence for internal use or for use by a third party.

(2)

Artificial intelligence developer— The term artificial intelligence developer means—

(A)

an entity that builds, designs, codes, produces, trains, or owns artificial intelligence for internal use or for use by a third party; and

(B)

does not include an entity that is solely a deployer of the artificial intelligence.

Sec. 203 Artificial intelligence questions in Federal surveys

(a)

In general— Not later than 1 year after the date of enactment of this Act, the Secretary of Commerce, acting through the Director of the Bureau of the Census, and the Secretary of Labor, acting through the Commissioner of Labor Statistics, shall, in consultation with the Chief Statistician of the United States, revise their respective surveys described in subsection (b) to incorporate or improve questions relating to the adoption and use of artificial intelligence in accordance with this section.

(b)

Surveys— The surveys described in this subsection are—

(1)

with respect to the Bureau of the Census—

(A)

the Annual Business Survey;

(B)

the Current Population Survey; and

(C)

the Business Trends and Outlook Survey; and

(2)

with respect to the Bureau of Labor Statistics—

(A)

the Occupational Requirements Survey; and

(B)

the American Time Use Survey.

(c)

Content— The questions and improvements required under subsection (a) shall—

(1)

to the extent practicable, be informed by the information collected under section 101 with respect to how best to implement this Act (including input or observations quantified or ranked in accordance with such section) and any other sources determined appropriate by the Secretary of Commerce or the Secretary of Labor; and

(2)

seek to obtain information on—

(A)

the types of artificial intelligence adopted by individuals or entities, to enable important analytical distinctions between such types, such as between language models and simple predictive models;

(B)

the occupations or tasks most impacted by artificial intelligence;

(C)

changes in skill requirements or training investments attributable to artificial intelligence, including shortages in skills necessary for the development, operation, or integration of artificial intelligence;

(D)

the intensity and frequency of artificial intelligence use; and

(E)

work outcomes attributable to artificial intelligence, such as changes in time spent or changes in quality or quantity of outputs.

(d)

Scope—

(1)

In general— Notwithstanding section 3, for purposes of the questions described in subsection (a), the Secretary of Commerce, or the Secretary of Labor, as either such Secretary determines appropriate for the purposes of this Act, may narrow the scope of artificial intelligence covered by such questions, such as by limiting the scope of such term to language models or by excluding from such term forms of artificial intelligence that do not represent the frontier of development or adoption of artificial intelligence.

(2)

Consultation— In developing the questions and improvements required by subsection (a), the Secretary of Commerce and the Secretary of Labor shall consult with each other regarding the scope of the term “artificial intelligence” used in such questions and, as appropriate and feasible, facilitate consistency.

(e)

Prohibition on additional appropriations— No additional amounts are authorized to be appropriated to carry out this section.

Sec. 204 Disclosures under the Worker Adjustment and Retraining Notification Act

Section 3 of the Worker Adjustment and Retraining Notification Act (29 U.S.C. 2102) is amended by adding at the end the following:

“(e) Artificial intelligence contributing to mass layoffs

“(1) In general—In the event that artificial intelligence was a substantial factor, as determined under the guidance issued under paragraph (3), in contributing to a mass layoff for which a notice is required under subsection (a), the notice required under such subsection shall include a statement—

“(A) specifying that artificial intelligence was a substantial factor in such mass layoff;

“(B) specifying the type and usage of such artificial intelligence;

“(C) providing an estimate of the percentage of the employment loss, with respect to such mass layoff, that is attributable to such artificial intelligence; and

“(D) specifying the actions, if any, that were taken to upskill or retrain employees to meet the needs of the employer prior to implementing such mass layoff.

“(2) Good-faith compliance—An employer satisfies the requirements under this subsection if the notice required under subsection (a) includes a good-faith statement with respect to the requirements under paragraph (1), including a good-faith estimate of the percentage required under subparagraph (C) of such paragraph.

“(3) Guidance—The Secretary of Labor shall issue guidance not later than 300 days after the date of enactment of the AI Workforce PREPARE Act on how employers may—

“(A) determine that artificial intelligence is a substantial factor in contributing to a mass layoff as described in paragraph (1); and

“(B) estimate the percentage required under subparagraph (C) of such paragraph.

“(4) Applicability—This subsection applies with respect to any notice issued under subsection (a) on or after the date that is 1 year after the date of enactment of the AI Workforce PREPARE Act.

“(5) Definition of artificial intelligence—In this subsection, the term artificial intelligence has the meaning given such term in section 3 of such Act.”