PRIA Logo™
119-hr-10594119th CongressWALLET

Reliable Artificial Intelligence Research Act of 2026

Sponsored By: Representative Pfluger, August [R-TX-11]

Introduced

Summary

Improve AI interpretability and adversarial robustness through DHS-run prize competitions. This bill would have the Department of Homeland Security set up prize challenges and coordinate with Commerce, NIST, NSF, the National Cyber Director, and other agencies to push practical tools, standards, and research for safer, more understandable AI systems.

Personalized for You

How does this bill affect your finances?

Personalize government policy and PRIA will tell you what this bill means for your household, plus every other piece of legislation we track. PRIA reads each provision against your financial profile to show you exactly what matters to your wallet.

Free to download

Bill Overview

Analyzed Economic Effects

3 provisions identified: 3 benefits, 0 costs, 0 mixed.

DHS report on contest results

This bill would require DHS to send a report within 180 days after the first prize competition ends. The report would go to the Senate Homeland Security and Governmental Affairs Committee and the House Homeland Security Committee. The report would explain how contest results inform interpretability and adversarial robustness, list gaps found, and recommend congressional actions to advance the research.

Definitions for AI testing terms

If enacted, this bill would define key AI terms used in the contests. "Adversarial robustness" would mean how well a model resists attacks while keeping integrity, reliability, and privacy. "Interpretability" would mean how well people can understand model decisions. "Red-teaming" would mean structured adversarial testing. "Artificial intelligence" would be defined by an existing federal law, and "Secretary" would mean the Secretary of Homeland Security.

DHS AI prize contests and funding

This bill would require DHS to start prize contests within 270 days after enactment. One contest would target adversarial robustness in a high-risk use. Another contest would target interpretability for widely used AI products. The contests could run in phases, use red-teaming, and DHS could hire outside partners. The bill would authorize $10 million for these contests for fiscal years 2027 through 2031, and Congress would need to appropriate the money.

Sponsors & CoSponsors

Sponsor

Pfluger, August [R-TX-11]

TX • R

Cosponsors

  • Rep. Harder, Josh [D-CA-9]

    CA • D

    Sponsored 9/24/2026

  • Rep. Newhouse, Dan [R-WA-4]

    WA • R

    Sponsored 10/9/2026

  • Del. King-Hinds, Kimberlyn [R-MP-At Large]

    MP • R

    Sponsored 9/24/2026

  • Rep. Downing, Troy [R-MT-2]

    MT • R

    Sponsored 9/24/2026

  • Rep. Webster, Daniel [R-FL-11]

    FL • R

    Sponsored 9/24/2026

  • Rep. Magaziner, Seth [D-RI-2]

    RI • D

    Sponsored 10/29/2026

  • Rep. Gottheimer, Josh [D-NJ-5]

    NJ • D

    Sponsored 10/22/2026

Roll Call Votes

No roll call votes available for this bill.

View on Congress.gov
Back to Legislation