Skip to content
WhoVotedWhy
119th Congress · S.1792In committeeCongress.gov ↗JSONReport an error

AI Whistleblower Protection Act

Summary, written by a model from the CRS summary

The AI Whistleblower Protection Act. Only the title was provided; it concerns whistleblower protections related to AI.

Model-rated AI relevance 90%: Whistleblower protections at AI companies are a direct part of AI governance. (claude-opus-5-5, Oct 22, 2026)

safetylabortransparency

Data as of: bills Oct 22, 2026 · votes Oct 21, 2026 · statements Oct 19, 2026. What changed

Progress

  1. Introduced

    May 15, 2025

  2. Reported by committee

    —

  3. Passed House

    —

  4. Passed Senate

    —

  5. Became law

    —

Latest action · May 15, 2025

Read twice and referred to the Committee on Health, Education, Labor, and Pensions.

What members said

Releases that cite this bill first, then releases on the same topics.

Christopher CoonsDemocrattoward oversightSep 14, 2026 · cites this bill by title

Coons calls for Congress to establish enforceable guardrails on AI risks and pass his AI Whistleblower Protection Act.

Source ↗
Sara JacobsDemocrattoward oversightOct 16, 2026 · same topic

Jacobs argues Congress should pursue both AI innovation and safety guardrails, addressing societal, model, deployer, and impact-level risks rather than prioritizing speed over safety.

Source ↗
Yvette ClarkeDemocrattoward oversightOct 15, 2026 · same topic

Clarke reintroduced legislation requiring disclosure when generative AI is used in political advertising to protect election transparency.

Source ↗
Pete RickettsRepublicanmixedOct 14, 2026 · same topic

Ricketts supports strengthening American AI innovation and export controls on chips to China while backing 'commonsense guardrails' for safety without stifling innovation.

Source ↗
Alexandria Ocasio-CortezDemocrattoward oversightOct 13, 2026 · same topic

Ocasio-Cortez and colleagues urge the FAA to pause its AI air traffic management tool until safety testing, oversight, and accountability questions are answered.

Source ↗
Lisa Blunt RochesterDemocratneutralOct 13, 2026 · same topic

Blunt Rochester describes meetings with AI experts and OpenAI leadership to discuss national security threats and recent AI safety issues, without stating a specific policy position.

Source ↗
Mark KellyDemocrattoward oversightOct 13, 2026 · same topic

Kelly advocates legislation requiring human oversight and accountability for military AI, ensuring servicemembers can operate when AI systems fail.

Source ↗
Tom BarrettRepublicantoward oversightOct 9, 2026 · same topic

Barrett introduced legislation creating a Cabinet-level federal office to oversee AI and emerging technology threats and ensure strong guardrails.

Source ↗

Bill data read 2026-10-22T14:48:01.167Z