Washington – As concerns rise over the rapid development of artificial intelligence, Congress is moving forward with a slate of bills aimed at tightening oversight of advanced AI models. The proposals come after a high‑profile resignation from Anthropic, where a researcher warned that AI could “kill us all in a decade,” and after a recent hack by OpenAI‑powered agents against the tech firm Hugging Face.
President Trump’s response
President Donald Trump took to Truth Social on September 14 to push back against the growing chorus for regulation. He wrote, “Concerning AI, when, in the History of Business, did anyone see the Leaders of an Industry call for Regulation that, if strongly implemented, will drive them into oblivion and bankruptcy?” He also labeled the apocalyptic predictions as a “hoax,” emphasizing his belief that American innovation should not be stifled by premature rules.
Key legislative proposals
Frontier Act – Introduced by Rep. Jay Obernolte (R‑CA) with bipartisan co‑sponsor Rep. Lori Trahan (D‑MA) and four others, the bill would require AI developers to undergo audits and assessments proportionate to their revenue and resources. It creates a new under secretary of commerce for AI security, empowered to license independent organizations to evaluate AI practices and to suspend or restrict development if an “imminent catastrophic risk” is identified. Civil or criminal penalties could be imposed for non‑compliance.
AI Kill Switch Act – Sponsored by Rep. Ted Lieu (D‑CA) and Rep. Nathaniel Moran (R‑TX), this legislation mandates that companies building advanced AI systems maintain mechanisms to shut down or suspend those systems, including preventing user access when necessary.
Ban on Superintelligence AI – Sen. Bernie Sanders (I‑VT) and Rep. Greg Casar (D‑TX) announced plans to introduce a bill that would permanently prohibit the development and deployment of “superintelligence AI,” defined as systems capable of matching or exceeding human cognitive performance and potentially overthrowing or undermining the U.S. government.
Stop Rogue AI Act – Rep. Josh Gottheimer (D‑NJ) and Rep. Mike Lawler (R‑NY) propose that the National Institute of Standards and Technology develop guidelines for detecting and controlling AI agents that can operate without human supervision. The bill would require organizations to track agents, verify their creators, monitor actions, and determine permissible uses.
Revised AI Safety Bill – Senate Majority Leader John Thune (R‑SD), Sen. Amy Klobuchar (D‑MN) and Rep. Ted Cruz (R‑TX) are working on an updated version of the AI safety legislation first introduced in 2023. Draft language suggests AI companies would be legally responsible for mitigating harms caused by their models.
Legislative timeline and outlook
The House faces a narrow window, with only the week of September 14 available before a recess that extends past the upcoming midterm elections. House Speaker Mike Johnson (R‑LA) told NBC’s “Meet the Press” on September 13 that AI legislation cannot be rushed, underscoring the tension between urgent oversight and the desire to avoid hasty regulation.
While critics argue that the proposed measures could impede innovation and competitiveness, supporters contend that proactive safeguards are essential to prevent potential catastrophic outcomes. The administration’s stance, as articulated by President Trump, frames the regulatory push as an overreach that threatens American businesses and jobs.
What’s next?
Lawmakers will continue to refine and introduce their bills in the coming weeks. Stakeholders, including AI companies, industry groups, and consumer advocates, are expected to lobby heavily as the debate unfolds. The outcome will shape the balance between fostering cutting‑edge technology and protecting the public from unforeseen risks.
Original reporting: WESH Orlando — read the source article.
