Republican Representative Nathaniel Moran of Texas introduced legislation on Thursday that would require AI model developers to report dangerous capabilities, security breaches, and safety incidents to the U.S. Commerce Department within seven days. The agency would then have 48 hours to notify Congress of the most serious incidents, creating a rapid federal alert system for frontier AI risks.
Reportable activities under the AI Incident Reporting Act include models attempting to evade human oversight, circumvent safeguards, or otherwise undermine the ability of human operators to control the system. Unauthorized access to model weights-the numerical parameters that guide a model's decision-making-and chemical, biological, nuclear, or other public safety threats also fall under the mandate. For legal teams advising AI developers, understanding these reporting thresholds is becoming a critical compliance function, and targeted training on AI for Legal can help build that expertise.
A narrow bill in a divided Congress
"It's a catch-it-early and sound-the-alarm bill," Moran said in an interview, signaling his intent to build a bipartisan coalition quickly. The targeted measure arrives as Congress struggles to pass broader AI legislation, with debates over whether federal law should preempt state regulations and whether guardrails would slow innovation and U.S. competition with China. Earlier this month, two House lawmakers released a discussion draft of the Great American Artificial Intelligence Act, which included similar incident reporting but took a wider regulatory approach.
Moran believes his more focused bill stands a better chance of becoming law. Mark Beall, president of the AI Policy Network, voiced support: "No legislation on AI has had much of a chance, but I think there's a growing demand from the public to see some action."
The urgency behind the bill intensified on June 12, when the Commerce Department directed Anthropic to disable global access to its latest models on national security grounds. That action exposed the absence of a transparent framework to govern frontier AI, a gap the legislation seeks to fill. The bill's emphasis on Commerce Department oversight reflects the government's expanding role in AI risk management, a focus area of AI for Government programs.
Why this matters for legal professionals
The proposed reporting requirements would create immediate compliance obligations for AI developers, and legal counsel will need to advise clients on what constitutes a reportable incident, the seven-day timeline, and the potential consequences of non-disclosure. The bill also adds a layer of federal oversight that could interact with state-level AI laws, raising preemption questions that lawyers will need to track. If the legislation passes, it will set a precedent for how the U.S. government formalizes incident-based AI regulation, making it essential for in-house and outside counsel at AI companies to understand the reporting framework and its triggers.
Your membership also unlocks:
