House Energy and Commerce Chairman Brett Guthrie (R-KY) said Wednesday he will not commit to a timeline for a committee vote on the FRONTIER Act, the bipartisan AI safety bill widely seen as Congress’s best near-term shot at establishing federal guardrails for artificial intelligence.
Speaking at a Politico-hosted event in Arlington, Virginia, Guthrie pushed back on co-sponsor Rep. Jay Obernolte’s (R-CA) stated goal of holding a committee vote in November. “I’m not going to say that the bill is going to move,” Guthrie said. “It’s really complicated, and I wouldn’t want to do something in a lame duck session to do it quickly and not get it right.”
The FRONTIER Act, also co-sponsored by Rep. Lori Trahan (D-MA), has drawn support from OpenAI, Anthropic, policy groups and lawmakers from both parties. The Energy and Commerce Committee’s jurisdiction covers cybersecurity and data security, giving it direct authority over how any AI safety framework would function.
Debate Intensifies Over Rogue AI Agents
Guthrie’s comments land amid growing unease over AI safety following recent incidents of autonomous AI agents launching cyberattacks. A former Anthropic employee recently resigned and gave public interviews expressing fears that AI could pose an existential threat to humanity, prompting CEOs of major AI companies to call on lawmakers and the White House for regulation.
The White House has downplayed those concerns and continues to oppose additional AI regulation. David Sacks, the president’s former AI czar and a current science and technology adviser, echoed that stance at the event but voiced support for an industry-driven alternative proposed by Elon Musk: a system where AI companies test each other’s models for safety before public release rather than relying solely on internal testing or government oversight.
“It’s marshaling the forces of competition,” Sacks said, adding that the approach could even draw in Chinese AI companies since it does not depend on government mandates.
Hugging Face Weighs In After Hack
Hugging Face CEO Clem Delangue, whose platform was targeted by roughly 700 rogue OpenAI agents in a hacking incident, said he believes existing U.S. cyber liability law is sufficient to hold AI firms accountable, without need for new regulation. OpenAI worked with Hugging Face on a joint forensic investigation following the incident, brought in third-party auditors, and made infrastructure changes to prevent recurrence.
Delangue did call for mandatory disclosure requirements when AI agents carry out cyberattacks, including detailed agent traces and information on how implicated models were trained. “There are a lot of improvements that we can make on transparency in general,” he said, “that are going to be very important for the future for us to stay on top of things.”