By Andy McDonald, SVP, Government Affairs, OAAA
The United States is currently operating within a fragmented regulatory framework for artificial intelligence, characterized by predominantly state-level activity, emerging federal coordination efforts, increasing litigation, and growing national security concerns shaping policy debates.
At the state level, AI regulation has accelerated dramatically. As of the date of this article, a majority of states have enacted laws addressing AI, with a smaller group—led by California, Colorado, Texas, Utah, New York, and Illinois – emerging as leading jurisdictions in developing cross-sector frameworks. These laws tend to focus on specific high-risk applications rather than comprehensive AI governance regimes, reflecting a targeted regulatory approach. Common areas include (i) employment and hiring tools, particularly algorithmic bias in automated decision-making; (ii) political deepfakes and election integrity protections; (iii) biometric data and facial recognition; (iv) government use of AI systems; and (v) consumer protection and deceptive practices. This patchwork has created significant compliance complexity for businesses operating nationally, raising concerns about inconsistent standards and regulatory arbitrage.
Simultaneously, constitutional and preemption challenges are emerging as a defining feature of the AI legal landscape. For example, Colorado’s original comprehensive AI law (SB 24-205) was challenged on First Amendment and other constitutional grounds, leading to a federal court order blocking enforcement. In May 2026, the state repealed and replaced the law with a narrower automated decision-making framework (SB 26-189), underscoring both legal vulnerability and legislative recalibration. Likewise, California’s political deepfake law (AB 2655) was struck down by a federal district court on Communications Decency Act preemption grounds, with the court declining to reach First Amendment issues; appellate review is ongoing. These cases signal that courts will play a central role in determining how far governments can regulate AI-generated content without infringing constitutional protections.
At the federal level, policymakers are attempting to balance innovation, competitiveness, and risk mitigation. In December 2025, a presidential Executive Order on artificial intelligence established a national policy favoring a minimally burdensome, innovation-driven framework and directed federal agencies to identify and challenge state laws that conflict with national objectives.
Policymakers have also considered proposals for a federal review framework governing cutting-edge “frontier” AI models prior to public release, including concepts that would allow agencies a defined evaluation period to assess cybersecurity and other systemic risks. While no such binding federal regime has been adopted to date, these discussions highlight internal divisions within the federal government between those favoring stronger pre-deployment oversight – particularly in light of AI-enabled cyber and national security risks – and those wary of slowing innovation or undermining U.S. competitiveness, especially relative to China. This dynamic illustrates a central policy tension: whether AI should be governed primarily through innovation-first frameworks or through more precautionary, national security–oriented controls on high-risk systems. It also suggests that future federal efforts may increasingly focus on frontier model governance, rather than only downstream applications.
A key legislative development is the “American Leadership in AI Act” (H.R. 8516), introduced in April 2026. This bill consolidates numerous prior proposals and emphasizes institution-building rather than direct regulation, including the creation of a Center for AI Standards and Innovation, the expansion of research infrastructure, and support for workforce development. Notably, it does not preempt state laws, reinforcing the current dual regulatory structure. The bill remains in early committee consideration and has not advanced toward enactment.
Another major regulatory frontier is AI infrastructure – particularly data centers, which are essential to AI deployment but raise significant legal and policy questions. Policymakers continue to grapple with energy consumption, environmental impacts, zoning and permitting challenges, and cost allocation. These issues reflect broader tensions between rapid AI expansion and sustainability, as well as resource constraints.
In sum, the current U.S. AI regulatory environment is defined not only by decentralization, constitutional uncertainty, and infrastructure pressures, but also by continued state activity and emerging federal indecision over how aggressively to regulate advanced AI systems at the model level. As national security concerns intensify, this debate is likely to play a central role in shaping the next phase of AI governance.
To receive a free morning newsletter with each day’s Billboard insider articles email info@billboardinsider.com with the word “Subscribe” in the title. Our newsletter is free and we don’t sell our subscriber list.
Paid Advertisement
