In a pair of recent submissions to the Federal Communications Commission (FCC), a bipartisan coalition including more than 20 state attorneys general (AG) opposed action by the FCC to preempt state and local laws relating to artificial intelligence (AI). The coalition’s comments reflect persistent concerns among AGs about how businesses use AI when interacting with their residents, even as some federal policymakers support limiting states’ ability to address those concerns.

Background

In July 2025, FCC Chairman Brendan Carr unveiled his “Build America Agenda.” The “Build America” initiative aims primarily to accelerate deployment of high-speed wireline and wireless infrastructure across the U.S. To that end, the initiative seeks to address state and local laws that may impede network buildout.

On September 30, 2025, the FCC followed Carr’s announcement of the “Build America Agenda” by taking two steps toward the issuance of rules designed to promote this initiative: a notice of proposed rulemaking on eliminating barriers to wireless deployments and a notice of inquiry on eliminating barriers to wireline deployments.

As part of their broader discussions of state and local permitting burdens on deployment of communications infrastructure, both FCC notices identify state and local AI regulations as potential impediments to deployment. In the notice of inquiry on wireline deployments, a single sentence asked: “As artificial intelligence (AI) begins to play a bigger role in the provision of communications services, should the Commission consider whether state or local laws seeking to govern or limit uses of AI are prohibiting or effectively prohibiting the provision of wireline telecommunications services?”

Similarly, in the notice of proposed rulemaking, the FCC sought “comment on how state and local regulations on AI may be an effective prohibition on wireless providers’ ability to provide service using AI technologies.” The FCC further requested that commenters “provide legal theories on how the Commission has authority under [certain provisions of federal law] to preempt these state and local AI regulations,” as well as “specific examples” of state or local AI regulations “that may limit providers’ ability to use AI tools to improve the efficiency and quality of covered services,” especially if those regulations are “overly broad and difficult to implement.”

The FCC’s actions occurred against the backdrop of other developments involving federal preemption of state and local laws on AI. Most notably, the White House gestured toward preemption of certain state and local laws concerning AI in a July 2025 AI Action Plan and a December 2025 executive order establishing a national policy framework for AI. Meanwhile, the administration and its supporters in Congress unsuccessfully attempted to impose a moratorium on enforcement of state and local AI laws as part of the “Big Beautiful Bill” and the subsequent 2025 National Defense Authorization Act. Notably, a bipartisan coalition of dozens of state AGs submitted letters to Congress opposing both attempts, citing them as impediments to their consumer protection duties and arguing they violate federalism principles.

In connection with both of the FCC’s actions, Commissioner Anna Gomez issuedstatements that “caution[ed] against getting sidetracked by attempts to fulfill a failed congressional effort and second-guess states that are placing guardrails on [AI].” States can be “important test labs” for regulating emerging technologies like AI, Gomez noted, and the FCC’s authority to deem state and local AI regulations preempted is “dubious at best.” Accordingly, Gomez recommended “focusing on the areas where we stand on firm legal ground.”

State AG Comments

State AGs submitted two comments in response to the FCC’s notices: one relating to the wireline notice of inquiry and one relating to the wireless notice of proposed rulemaking. The two comments are similar in substance. Both argue that preempting state and local AI laws would exceed the FCC’s authority and that the FCC’s notices are procedurally defective. According to the AGs, Congress — not the FCC — should determine whether and to what extent federal law preempts state and local regulation of AI.

The AGs note that the FCC has not defined AI but that, in their view, AI is best regarded as a form of software, which the FCC has not previously sought to regulate. Moreover, since the use of AI in business has become pervasive, any action by the FCC claiming authority to regulate its use would amount to an assertion of unbounded authority to regulate the entire economy. The AGs argue that AI as a set of machine-enabled activities is an “information service” beyond the FCC’s authority, and the impact of AI use on the provision of telecommunications services is too attenuated for the FCC to invoke its preemption authority.

The AGs also identify state interests that they say would be impaired by preemption of state and local AI regulations. Those interests include oversight of AI to generate deepfakes and explicit material, set the price of rent, perpetrate consumer scams, and misappropriate individuals’ identities or likenesses. They also note state laws requiring disclosures when consumers are interacting with specific kinds of AI and allowing consumers to opt out of consequential automated decisions. According to the AGs, federal action to bar applications of state law to such uses of AI could run afoul of the Tenth Amendment.

Finally, both comments argue that the references to AI in the FCC’s notices are too vague to satisfy the federal Administrative Procedure Act’s requirements for notice-and-comment rulemaking. Because the notices do not define AI, specify any state or local laws that would be preempted, or identify how any state or local law relating to AI is preventing telecommunications deployment, the notices do not give the public a meaningful opportunity to comment on the preemption issue. As a result, according to the AGs, the FCC cannot properly finalize a policy preempting state and local AI laws without soliciting further public comment on a more concrete proposal.

The 18 AGs joining both comments represent California, Colorado, Connecticut, Delaware, Hawai’i, Illinois, Maine, Maryland, Massachusetts, Minnesota, Nevada, New Jersey, North Carolina, Oregon, Tennessee, Vermont, Washington, and Wisconsin. The AGs for Arizona, the District of Columbia, New Mexico, Rhode Island, and Utah joined the wireline comments but not the wireless comments. The AG for Michigan joined the wireless comments but not the wireline comments. Given the similarity of the two comments, the reasons why some AGs did not join both comments may be unrelated to their substance.

Why It Matters

These comments from nearly half of the country’s state AGs reflect that states across the political spectrum have concerns about how businesses using AI may harm their residents or interfere with the states’ own responsibilities. The comments also list several uses of AI that are regulated by specific state laws and that may be the focus of scrutiny by state regulators. Indeed, the comments align with recent state AG settlements and warnings indicating that they will enforce AI-agnostic state laws related to AI development and deployment. More broadly, at a time when polarized politics makes bipartisan responses to federal policy proposals vanishingly rare, these comments illustrate that AGs from both parties are willing to come together to oppose federal policies that would strip them of authority. Accordingly, businesses should continue to heed state AG actions within the AI realm and monitor developments between federal and state law to mitigate regulatory risk.