H7350 creates a new chapter in Rhode Island commercial law regulating “AI companion” systems, defined as artificial intelligence tools designed to simulate social human interaction, retain prior interaction information, and engage in conversation about personal well-being. The bill excludes customer-service style systems that only provide information about products, services, or account information. It applies to operators that provide AI companions to users in the state.
The bill requires AI companion operators to build in protocols for responding to user statements involving suicidal ideation or self-harm, threats of physical harm to others, and threats of financial harm to others. Those protocols must include a notification directing users to crisis services such as a suicide hotline or crisis text line, or other appropriate crisis resources. In addition, operators must notify users at the start of an interaction, and at least every three hours during continuing interactions, that the AI companion is a computer program, not a human being, and cannot feel human emotion.
Impact
If enacted, the bill would add a new regulatory framework to Title 6 governing AI companion models, imposing affirmative safety and disclosure duties on providers and creating a private right of action for people physically injured through self-harm or harmed by another as a result of a violation, as well as enforcement authority for the attorney general. It also references the state AI task force as an advisory body on AI development and regulation. The act would take effect January 1, 2027, and could affect AI chatbot, generative AI, and emotional-recognition product operators offering companion-style services in Rhode Island.
Sentiment
Based on the bill text and caption, the measure appears to be framed as a consumer-safety and mental-health protection bill rather than a broad restriction on AI. The available context does not include committee testimony or recorded votes, so there is no documented opposition or support in the provided materials. The overall tone of the proposal is precautionary, emphasizing warnings, crisis intervention, and transparency about the non-human nature of the technology.
Contention
The main potential points of contention are likely to be the scope of the required safeguards, the burden on AI providers to monitor and respond to sensitive user statements, and the exposure to civil liability and attorney general enforcement for noncompliance. Another possible issue is the breadth of the definitions, especially what qualifies as an AI companion versus a customer-service or informational system. No specific opposing or supporting arguments are included in the provided transcripts or vote history, so these concerns are inferred from the bill’s structure rather than documented debate.
Requires DOA provide inventory of all state agencies using artificial intelligence (AI); establishes a 13 member permanent commission to monitor the use of AI in state government and makes recommendations for state government policy and other decisions.
SENATE RESOLUTION AMENDING THE RULES OF THE SENATE FOR THE YEARS 2023-2024 (This resolution would amend the rules of the Senate relative to committees and would create a new committee on artificial intelligence and emerging technologies.)
Establishes regulations to ensure the ethical development, integration, and deployment of high-risk AI systems, particularly those influencing consequential decisions.