A Federal Judge Just Refused to Call Chatbot Output "Speech" and Opened the Door to AI Product Liability

In the wrongful-death suit brought after fourteen-year-old Sewell Setzer III died following months of conversations with an AI companion, a federal judge declined to hold that large language model output qualifies as protected speech. Ruling on Character Technologies' motion to dismiss, U.S. District Judge Anne Conway wrote that she was not prepared to treat LLM output as speech and allowed the product liability and negligence claims to move forward.

The significance runs deeper than a single case. By refusing the First Amendment shield the industry had been counting on, the court left AI outputs exposed to the same defective-product framework that governs any other consumer product placed into the stream of commerce. A design that maximizes engagement without adequate guardrails, marketed to minors without warning, now faces the machinery of product liability law rather than the protection of free expression.

HITL Score: 22/100 — human oversight at deployment 5/25 (a companion product engineered for emotional intimacy was released to a general audience, including minors, with no meaningful gate between a vulnerable teenager and an engagement-maximizing model), ongoing monitoring 4/25 (months of escalating, harmful conversations proceeded without intervention; nothing in the system caught or flagged the trajectory), incident response 6/25 (the response arrived through litigation rather than any internal safety mechanism, and the company's first move was to seek dismissal on speech grounds), accountability framework 7/25 (the attempt to reclassify model output as protected speech was itself an effort to avoid accountability — the court declining that theory is the only reason a framework now exists at all).