Skip to main content
S2195AaaTechnology

Commercial Law - General Regulatory Provisions - Artificial Intelligence Companion Models

This bill requires AI companion apps to include crisis protocols for users expressing self-harm and mandates non-human disclosure warnings.

Introduced36 Yea0 Nay2 Not voting
Population
Affected
25
Email Legislator
Introduced Jan 23, 2026Committee Senate Artificial Intelligence & Emerging Tech

Plain-English Summary

This bill requires companies that provide artificial intelligence (AI) companions—programs designed to simulate a personal relationship with a user—to include specific safety features. AI companions must have protocols to detect if a user expresses thoughts of self-harm, suicide, or harming others, and must provide the user with contact information for crisis services, like a suicide hotline. Companies must also remind users at the start of a conversation and every three hours that they are talking to an AI, not a real person. The Attorney General can fine companies up to $15,000 per day for violations.

For younger readers

This bill makes rules for computer programs called AI companions. These are computer characters that talk to you like a friend. The bill says these programs must have a special plan if someone tells the computer they want to hurt themselves or someone else. If that happens, the computer must give the person a phone number to call for help. Also, the computer program must remind the person when they first start talking, and every three hours after, that they are talking to a computer and not a real, live human being.

Who & Where It Applies

Impacted groups
AI companion operatorsAI companion usersMental health crisis providersTechnology companiesOffice of the Attorney General
Impacted communities
All

Constitutional & Fiscal Check

There is a slight First Amendment risk regarding compelled speech. The bill mandates specific disclosures (notifying users they are talking to an AI every three hours) and compelled notifications referring users to crisis services. While commercial speech can be regulated for consumer protection, tech companies might argue these mandates are overly prescriptive and infringe on their editorial control over generative AI outputs. However, as consumer protection warnings, they are likely constitutional.

Estimated cost
None
Estimated revenue
Up to $15,000/day per violation for noncompliant operators (directed to suicide prevention programs)

Bill Analysis

Both viewpoints
For Progressives
  • Promotes the welfare of the community by requiring mental health crisis interventions and directing users to suicide prevention resources when they are vulnerable.
  • Directs corporate penalty fines toward funding suicide prevention programs, expanding public health resources without taxing the public.
  • Enacts strong consumer protection and transparency rules by forcing tech companies to disclose that users are speaking to a machine, protecting vulnerable individuals from emotional manipulation.
  • Relies on AI models to accurately detect emotional distress or self-harm, which could lead to biased, inaccurate, or culturally insensitive automated responses that fail marginalized groups.
  • Mandating data reporting to the Attorney General regarding crisis protocol activations may raise concerns about government surveillance and the privacy of vulnerable individuals experiencing mental health crises.
  • Does not provide direct state funding for the crisis centers and suicide hotlines that will likely see an increase in call volume due to these automated referrals.
For Conservatives
  • Holds technology companies accountable for their products by ensuring they clearly disclose when consumers are interacting with artificial intelligence rather than a real human.
  • Protects traditional social values and human relationships by placing boundaries on artificial companions that attempt to mimic authentic human emotional connections.
  • Funds suicide prevention programs through penalties on violating corporations rather than relying on broad taxpayer-funded government assistance programs.
  • Imposes heavy regulatory burdens and government mandates on technology companies, interfering with corporate freedom and free market product design.
  • Empowers the Attorney General with broad authority to investigate and sue businesses, potentially weaponizing state power against specific tech companies.
  • Establishes severe civil penalties of up to $15,000 per day, which could cripple smaller tech startups and deter artificial intelligence innovation and investment in the state.

Votes

Loading votes...

Full Bill Text

3 versions
Changes to existing Rhode Island law · 64 additions

SECTION 1. Title 6 of the General Laws entitled "COMMERCIAL LAW — GENERAL REGULATORY PROVISIONS" is hereby amended by adding thereto the following chapter: 6-63-1. Definitions. As used in this chapter: (1)(i) "AI companion" means a system using artificial intelligence, generative artificial intelligence, and/or emotional recognition algorithms to simulate a sustained human or human-like relationship with a user by:

(A) Retaining information on prior interactions or user sessions and user preferences to personalize the interaction and facilitate ongoing engagement with the AI companion;

(B) Asking unprompted or unsolicited emotion-based questions that go beyond a direct response to a user prompt; and

(C) Sustaining an ongoing dialogue concerning matters personal to the user. (ii)"AI companion" shall not include:

(A) Any system used by a business entity solely for customer service or to strictly provide users with information about available commercial services or products provided by such entity, customer service account information, or other information strictly related to its customer service;

(B) Any system that is primarily designed and marketed for providing efficiency improvements or research or technical assistance; or

(C) Any system used by a business entity solely for internal purposes or employee productivity.

(2) "AI model" means a component of an information system that implements artificial intelligence technology and uses computational, statistical, or machine-learning techniques to produce outputs from a given set of inputs.

(3) "Artificial intelligence", "artificial intelligence technology", or "AI" means a machine- based system that can, for a given set of human-defined objectives, make predictions, recommendations, or decisions influencing real or virtual environments.

(4) "Emotional recognition algorithms" means artificial intelligence that detects and interprets human emotional signals in text (using natural language processing and sentiment analysis), audio (using voice emotion AI), video (using facial movement analysis, gait analysis, or physiological signals), or a combination thereof.

(5) "Generative artificial intelligence" means a class of AI models that are self-supervised and emulate the structure and characteristics of input data to generate derived synthetic content including, but not limited to, images, videos, audio, text, and other digital content.

(6) "Operator" means any person, partnership, association, firm, or business entity, or any member, affiliate, subsidiary or beneficial owner of any partnership, association, firm, or business entity who operates or provides an AI companion.

(7) "Person" means any natural person.

(8) "User" means any person who uses an AI companion within the state and who is not an operator or agent or affiliate of the operator of the AI companion. 6-63-2. Prohibitions and requirements.

(a) It shall be unlawful for any operator to operate or provide an AI companion to a user unless such AI companion contains a protocol for addressing:

(1) Possible suicidal ideation or self-harm expressed by a user to the AI companion;

(2) Possible physical harm to others expressed by a user to the AI companion; and

(3) When any of those expressions are made, a notification shall be provided to the user that refers them to crisis service providers such as a suicide hotline, crisis text line, or other appropriate crisis services as soon as any of those expressions are detected.

(b) Beginning July 1, 2027, operators shall file annual reports with the office of the attorney general which shall include the number of safety protocol activations and related metrics. The office of the attorney general shall publish aggregated data on its website. 6-63-3. Notifications required. An operator shall provide a clear and conspicuous notification to a user at the beginning of any AI companion interaction and at least every three (3) hours for continuing AI companion interactions hereafter, which states either verbally or in writing that the user is not communicating with a human. 6-63-4. Enforcement.

(a) The attorney general shall be empowered to enforce this chapter, including the power to investigate, sue, and seek injunctions against noncompliant AI companion operators for violations of this chapter.

(b) Violations of this chapter may result in civil penalties up to fifteen thousand dollars ($15,000) per day, with fines directed to suicide prevention programs. 6-63-5. Severability. If any provision of this chapter is declared unconstitutional or invalid, or the application of any provision of this chapter to any person or circumstance is held invalid, the constitutionality of the remainder of the chapter and its applicability to other persons and circumstances shall not be affected thereby.

SECTION 2. This act shall take effect on January 1, 2027.