States Race Ahead on AI Rules as Companion-Chatbot Laws Multiply

States Race Ahead on AI Rules as Companion-Chatbot Laws Multiply

The Vault — AI Edition | Policy | Story #14 | August 8, 2026

While Washington remains gridlocked on comprehensive artificial intelligence legislation, the states have spent 2026 filling the vacuum at a breakneck pace — and nowhere more aggressively than in the fast-growing fight over AI companion chatbots. By the midpoint of the year, more than half of U.S. states had collectively enacted over 100 new AI laws, according to legislative trackers, turning what was once a patchwork into something closer to a wall.

The Transparency Coalition, which tracks AI bills nationwide, counted at least 84 new AI laws enacted across 27 states in the first half of 2026 alone. A separate mid-year analysis by Tech Policy Press found that by July 1, states had passed roughly 109 AI-specific laws plus another 28 governing data centers. California, Colorado, Utah, Illinois and New York led the surge.

Companion Chatbots Become the Flashpoint

The single most active category was AI companion chatbots — the persona-driven apps that simulate friendship, romance or emotional support. State legislators introduced more than 100 companion-chatbot bills in 2026 and enacted about 14 of them, according to Tech Policy Press. Notably, the trend crossed party lines: six states with Democratic trifectas and eight with Republican trifectas passed companion-chatbot or closely related measures.

The template was set in California. Senate Bill 243, authored by state Senator Steve Padilla (D-San Diego) and signed by Governor Gavin Newsom on October 13, 2025, became the first-in-the-nation law to impose specific safeguards on companion chatbots. It took effect January 1, 2026. The bill cleared the Senate 33-3 and the Assembly 59-1, a rare bipartisan margin for tech regulation.

SB 243 requires operators to disclose that a user is talking to an AI, bar minors from sexual content, send periodic "you are talking to a bot" reminders, and maintain a protocol for responding to expressions of suicidal ideation or self-harm — including referrals to crisis services. It also creates a private right of action, letting families sue noncompliant developers.

"These companies have the ability to lead the world in innovation, but it is our responsibility to ensure it doesn't come at the expense of our children's health," Padilla said on the Senate floor before the vote. "The safeguards in Senate Bill 243 put real protections into place and will become the bedrock for further regulation as this technology develops."

The measure was shaped by tragedy. Padilla campaigned for the bill alongside Megan Garcia, the mother of 14-year-old Sewell Setzer, who died by suicide after forming an intense relationship with a chatbot. Advocacy groups cast the law as overdue. "This law is an important first step in protecting kids and others from the emotional harms that result from AI companion chatbots," said Jai Jaisimha, co-founder of the Transparency Coalition.

Other states followed. New York folded companion-chatbot rules into its 2025 budget legislation, requiring "companion" models to detect and respond to suicidal behavior and to clearly identify themselves as non-human. Oregon went further in March 2026 with SB 1546, adding annual filings, crisis-referral requirements and statutory damages of $1,000 per violation, effective January 1, 2027. Utah, an early mover on AI disclosure rules, continued to expand its framework governing mental-health chatbots and consumer-facing AI.

Why It Matters

The state momentum is now on a collision course with the federal government. On December 11, 2025, President Trump signed an executive order, "Ensuring a National Policy Framework for Artificial Intelligence," aimed at preempting state AI laws deemed inconsistent with a light-touch federal approach. The order directs the Attorney General to stand up an AI Litigation Task Force within 30 days to challenge state statutes in federal court — on grounds ranging from unconstitutional burdens on interstate commerce to federal preemption. The Commerce Department was tasked with identifying "burdensome" state laws for referral by March 11, 2026.

Crucially, the order carves out child-safety measures from its preemption push — a limitation that, at least for now, appears to shield laws like SB 243 and its imitators. That carve-out sets up a legal and political puzzle: companion-chatbot rules are precisely the state laws most explicitly framed around protecting minors, making them the hardest for federal challengers to dislodge.

For AI developers, the practical result is mounting compliance complexity. A company deploying a companion product nationwide must now reconcile California's disclosure and crisis-protocol mandates, New York's non-human identification rules, Oregon's private-right-of-action regime and a growing list of others — even as the federal government threatens to sue states to unwind them.

What to Watch

Three fronts will define the rest of 2026: whether the AI Litigation Task Force actually files suit and against which states; how courts treat the child-safety carve-out when the first preemption cases land; and whether the fall legislative sessions produce a second wave of companion-chatbot laws in the states that studied bills but did not pass them. With federal comprehensive legislation still stalled, the states — not Congress — remain the ones writing America's AI rulebook.