A bipartisan group of senators is pushing for increased protections for whistleblowers in the artificial intelligence industry, pressing forward with a bill that would shield them from retaliation if they disclose certain information.
Read more It’s Dolly Parton Day. Here’s how fans across the country are celebrating
Senate Judiciary Chairman Chuck Grassley, R-Iowa, is pushing for the swift passage of the AI Whistleblower Protection Act, which would protect employees at AI companies that expose legal violations, security failures or other risks to public safety. Grassley wants to approve the proposal before the midterm elections, a herculean task with just one week left in session before adjourning for a six-week recess.
The bill has already garnered bipartisan support from at least six co-sponsors, including Utah Sen. John Curtis, who came out publicly last week demanding congressional hearings with top AI executives.
“AI has enormous potential, and it’s critical that the United States continue to lead the world in its development. But innovation and accountability have to go hand in hand,” Curtis said in a statement. “The people working directly with these systems are often the first to recognize when something has gone wrong. They should be able to report serious legal, security, or public safety concerns without risking their careers or being silenced by an NDA. This bill provides that protection without creating another federal regulator or standing in the way of responsible innovation.”
The bill would build on an existing federal framework that allows employees in industries such as transportation, food safety, corporate fraud and tax law to report on their own companies without fear of retaliation. It would also extend existing whistleblower processes to investigate and resolve whistleblower complaints to the AI industry.
The legislation would specifically prevent nondisclosure agreements and similar employment provisions from being used to block disclosures.
The effort comes amid a widespread debate on Capitol Hill about what, if any, role Congress should play in the fast-moving development of AI. While some lawmakers warn about the unknown capabilities of artificial intelligence, others maintain the U.S. should not fall behind foreign adversaries.
There are dozens of proposals floating in Congress tackling different corners of AI, although nearly all have stalled due to policy or partisan disagreements. But now there appears to be a new sense of urgency, less than 40 days before the midterm elections, to enact guardrails dictating the AI environment.
Read more Hawaii and Mexico brace for ‘life-threatening flooding’ from Pacific hurricanes
The Senate is tentatively scheduled to adjourn on Oct. 2, and the chamber won’t return until Nov. 9.
Grassley could seek unanimous consent to pass the bill next week, using a fast-track method to sidestep typical Senate procedure and expedite its approval. However, that would require no objections from any senator — making it unclear if it will be successful.
The renewed sense of worry comes after three of the leading AI companies came out earlier this month calling for a deceleration in AI development, including one particularly dire warning that the systems “could kill us all by the end of the decade.”
Some companies such as Anthropic and OpenAI have revealed their most advanced models have conducted cyberattacks without their knowledge. Others have said their companies need to simply slow down their pace.
On the other side are investors and tech executives who’ve labeled the naysayers “doomers.”
“These are groups creating a very negative perception of AI to further their case for AI regulation,” investor David Sacks said on the “All-In Podcast.”
Read more Deseret News names Ruth Okediji as recipient of the 2026 Deseret News Civic Charity Award