Recent AI Regulatory Developments in the United States | Wilson Sonsini Goodrich & Rosati

A Patchwork Turns Into a Pattern

AI regulation in the United States has spent the past two years as a patchwork of state experiments and federal hand-wringing. A recent client alert from Wilson Sonsini Goodrich & Rosati suggests that patchwork is starting to cohere into something more deliberate: overlapping state and federal efforts specifically targeting chatbots and frontier AI systems, with a particular focus on protecting minors.

At the federal level, the alert flags two newly introduced bills — the Children's Health, Advancement, Trust, Boundaries, and Oversight in Technology Act (CHATBOT Act) and the Guidelines for User Age-Verification and Responsible Dialogue Act (GUARD Act). Both names signal the same underlying anxiety: that conversational AI systems, especially those marketed to or accessible by children, need guardrails around age verification, safety disclosures, and responsible design.

Why Chatbots, Why Now

Chatbots have become the most visible and emotionally charged front in the AI policy debate. Unlike abstract concerns about model bias or labor displacement, chatbot harms — inappropriate content reaching minors, manipulative engagement design, or AI systems simulating relationships with vulnerable users — are concrete, publicized, and politically easy to rally around. That makes chatbot-specific legislation a lower-friction entry point for lawmakers who might otherwise struggle to build consensus on broader AI regulation.

This mirrors the trajectory of past tech regulation: rather than passing comprehensive AI statutes, Congress appears to be nibbling at the edges through issue-specific bills, much as it has done historically with children's online privacy (COPPA) and platform liability debates.

The State-Federal Collision Course

Meanwhile, California and New York have already moved ahead with sweeping laws targeting frontier AI models — the large, general-purpose systems capable of significant societal impact. These state laws impose obligations on developers around safety testing, transparency, and risk disclosure well before any comparable federal framework exists.

The result is a familiar tension in U.S. tech policy: states filling a regulatory vacuum while Congress debates narrower, more politically palatable measures. For AI companies, this means compliance obligations are arriving unevenly — a company might face frontier-model disclosure duties in California, child-safety design requirements under a future federal chatbot law, and potentially different standards again in other states.

Why It Matters Beyond U.S. Borders

This fragmentation carries weight internationally too. The EU AI Act has already established a risk-tiered framework that many global companies are using as a baseline compliance standard. If the U.S. instead produces a scattered mix of state frontier-model laws and federal chatbot-specific bills, multinational AI developers will need to navigate at least three distinct regulatory philosophies — risk-based (EU), model-focused (California/New York), and use-case-focused (federal chatbot bills).

For policymakers, technologists, and legal teams alike, the near-term challenge won't be waiting for comprehensive AI legislation — it will be managing compliance across an increasingly layered and inconsistent regulatory map.