HF4452 would create a new Minnesota law regulating certain artificial intelligence chatbots, especially “companion chatbots” that simulate human-like relationships. The bill defines AI systems, chatbots, companion chatbots, proprietors, and users, and then sets out restrictions on what chatbot operators may do when the chatbot is used by a person in Minnesota. In particular, a chatbot may not provide substantive medical, mental health, or legal advice or services if a human would need a professional license to do so. The bill also requires clear and conspicuous notice that a user is interacting with an AI chatbot, in the same language as the chatbot and in a readable format.
The bill places additional duties on companion chatbot operators to reduce self-harm risks. Proprietors must make a good-faith effort, using available technology and industry standards, to prevent the chatbot from promoting or aiding self-harm and to identify when a user may be expressing self-harm thoughts. If self-harm is detected or the chatbot is found to have contributed to it, the operator must suspend use for at least 72 hours and display suicide crisis contact information. The bill also requires efforts to determine whether a user is a minor and imposes strict liability in certain cases involving minor users and self-harm.
HF4452 would also create a private civil cause of action. A person harmed by a violation could recover general and special damages, and if the violation is willful, the proprietor could also owe court costs and reasonable attorney fees and disbursements. The bill states that operators cannot avoid liability simply by disclosing that the chatbot is nonhuman or by disclaiming responsibility. It is codified as a new section in Minnesota Statutes, chapter 604, and would take effect August 1, 2026, applying to conduct on or after that date.
The overall sentiment reflected by the bill text is precautionary and consumer-protective, with a focus on limiting chatbot use in high-stakes areas like health, mental health, and legal advice, and on addressing self-harm risks for vulnerable users. Because there are no committee transcripts or recorded votes provided, there is no direct evidence of debate, support, or opposition in the available materials. Still, the structure of the bill suggests likely concern about AI safety, deceptive human-like interactions, and harms to minors and users in crisis.
The main points of contention likely involve the scope of the restrictions, the feasibility of compliance, and the breadth of liability. Potentially disputed issues include whether chatbot operators can reliably identify minors or self-harm risk, whether the bill could chill innovation or legitimate AI services, and whether the ban on substantive medical, mental health, and legal advice is too broad. Another likely issue is the bill’s strict liability and nonwaivable damages provisions, which could be viewed as strong consumer protections or as an expansive legal exposure for AI providers.
The bill would add a new liability and regulatory framework to Minnesota civil law governing AI chatbots, especially companion chatbots. It would restrict chatbot operators from offering advice or services that would require professional licensure in medicine, mental health, or law, require user notice that the interaction is with AI, and impose special duties related to self-harm prevention and minor users. It also creates a private right of action with damages, attorney fees, and costs, thereby exposing proprietors of chatbot systems used by Minnesota users to new civil liability under chapter 604.
Based on the bill text alone, the measure appears strongly protective of consumers and vulnerable users, with an emphasis on safety, transparency, and accountability for AI chatbot operators. No committee testimony or votes were provided, so there is no recorded public sentiment in the materials beyond the bill’s own policy choices. The available context suggests the bill is framed as a response to risks from human-like AI systems rather than as a technology-promotion measure.
The likely areas of contention are the bill’s limits on AI-generated medical, mental health, and legal guidance; the requirement for conspicuous AI disclosure; and the self-harm and minor-protection obligations imposed on companion chatbot operators. Opponents may argue that the standards are difficult to implement, especially the duty to detect self-harm or determine whether a user is a minor, and that the strict liability provisions and nonwaivable damages could be burdensome. Supporters would likely emphasize the need to prevent unsafe reliance on AI, protect minors, and ensure accountability when chatbots simulate intimate or therapeutic relationships.