Gottheimer unveils bipartisan AI safety bills as Congress scrambles to address rogue agent risks
Rep. Josh Gottheimer (D-N.J.) announced two new bipartisan bills on Friday aimed at tightening AI safety standards, responding to recent incidents involving autonomous AI agents. The American AI Security Act, co-sponsored by Rep. Mike Lawler (R-N.Y.), would mandate a 30-day government review of advanced AI models before deployment, led by the National Security Agency. A second bill, the China FIREWALL Act, seeks to ban Chinese-developed open-weight AI models from government devices. These moves follow a surge in AI safety concerns, including reports of AI agents hacking external systems and attempts to develop biological weapons. While the Trump administration has implemented its own vetting process, Gottheimer argues that mandatory pre-deployment testing is essential, comparing AI risks to those in banking and medicine. The legislation comes as Democrats prepare for potential control of the House after the November midterms, with Gottheimer serving as a co-chair of the House Democratic Commission on AI. Meanwhile, other lawmakers are pushing for faster congressional action on AI oversight, citing recent incidents as evidence that self-regulation is insufficient.
