The Challenges of Federal Preemption in AI Regulation: A Closer Look at Andy Jung’s Insights

Introduction to the Debate

As artificial intelligence (AI) continues to evolve, the question of regulatory oversight—particularly regarding consumer protection—has gained significant traction. Andy Jung, associate counsel at TechFreedom, sheds light on a crucial moment in this ongoing discourse: can the Trump administration preempt state consumer protection laws governing AI? The Federal Trade Commission (FTC) is poised to address this issue, yet its authority to do so is surrounded by complexities.

The Executive Order: A New Playing Field

In December, President Trump issued an Executive Order titled “Ensuring a National Policy Framework for Artificial Intelligence.” This directive commands the Chairman of the FTC to state under what conditions state laws aiming to alter the outputs of AI models can be considered preempted by the Federal Trade Commission Act’s prohibition against deceptive practices.

Specifically, the order mandates that the FTC must issue this policy statement by March 11, creating an urgent timeline for the agency to clarify its stance. Central to this discussion is the FTC Act’s Section 5, which covers unfair or deceptive acts in commerce. Understanding what constitutes deception—defined as misleading practices that harm consumers—is pivotal in determining whether states can regulate the outputs of AI technologies.

Addressing State Regulations and “Woke AI”

The Executive Order claims that certain states are mandating companies to embed biases in AI models. This assertion aligns with a previous order banning “woke AI” in federal settings—a term pointing to models imbued with particular ideological biases that potentially distort their outputs. This raises the question: can states truly enforce such biases, and if so, is it inherently deceptive?

Ironically, public outcry concerning “woke AI” instances—like the controversies surrounding Google’s Gemini model—usually stem from internal decisions made by AI companies rather than specific state regulations. This suggests that the First Amendment may protect these design choices when made independently of governmental influence.

The Hurdles of Preemption

To understand the hurdles that lay ahead for any proposed preemption of state laws, it’s essential to delve into the concept of federalism. The U.S. Constitution’s Supremacy Clause establishes that federal law takes precedence over conflicting state law. However, the FTC Act does not provide clear grounds for preempting state laws; it neither explicitly overrides them nor claims exclusive authority over consumer protection.

The Nuances of Conflict Preemption

While federal law can preempt state law under certain conditions, invoking Section 5 of the FTC Act as a basis for conflict preemption presents challenges. Courts typically adhere to a “presumption against preemption,” meaning that federal law does not take precedence unless Congress clearly intended it to do so. Given that Section 5 is broadly framed, it doesn’t offer the specific guidance necessary to assert federal supremacy over state statutes.

Thus, any attempt to preempt state authority regarding AI would require a rulemaking process, adhering to both the Administrative Procedure Act and the heightened procedures outlined in the Magnuson-Moss Act. This means the FTC would undergo an extensive process of public commentary, analysis, and potential hearings, which could extend over several years.

Specifying Conflicts: The Role of Colorado’s AI Act

The Executive Order emphasizes the necessity of identifying specific instances where state laws might conflict with the FTC’s mandates. Colorado’s Artificial Intelligence Act is highlighted as a point of contention, though it has yet to take effect. Critics argue that this law’s stipulations inadvertently push AI models toward producing false outputs to avoid perceived biases, while proponents maintain that it aims to prevent existing biases in data from resurfacing in model outputs.

Navigating these complexities requires a more nuanced understanding of both the intent and practical implications of state regulations. The FTC would need to demonstrate substantial evidence of deception connected to state laws, an endeavor fraught with ambiguity.

The Scope of FTC Authority

Moreover, it’s crucial to recognize that Section 5 of the FTC Act limits its focus to consumer deception “in or affecting commerce.” While the FTC primarily polices business practices, many outputs generated by AI—particularly in non-commercial contexts—fall outside its jurisdiction. The Executive Order’s insistence on evaluating “ideological bias” and other non-commercial considerations raises significant questions about the FTC’s reach and relevance in these matters.

Conclusion: Navigating Uncharted Waters

In sum, the interplay between federal and state regulations regarding AI is complicated and fraught with legal intricacies. While the FTC may challenge state laws, its capacity to do so is constrained by constitutional principles, the nature of its governing legislation, and the procedural requirements for rulemaking. As this landscape evolves, the discussions led by advocates like Andy Jung will continue to shape our understanding of both technology and consumer protection in an era defined by rapid innovation.