AI companions must give crisis referrals starting January 2027

Operators must equip covered AI companions with a plan to address users expressing thoughts of suicide, self-harm, or physically harming others. The rules cover operators serving people who use AI companions in Rhode Island. Operators and their agents and affiliates do not count as users. Covered companions remember earlier interactions and preferences, ask unprompted emotional questions, and sustain personal conversations. The law excludes tools used solely for customer service or internal work, and tools primarily designed and marketed for efficiency, research, or technical help. The safety requirements take effect January 1, 2027.

Operators must clearly tell users, verbally or in writing, that they are not communicating with a human. The notice applies to interactions with covered AI companions used by people in Rhode Island. Operators must give the notice at the beginning of each interaction and at least every three hours while it continues.

Operators must report safety response activations annually to the Rhode Island attorney general, who must publish combined data and can enforce the law. These duties and penalties apply to operators of covered AI companions. Reports must include activation counts and related measures. Violations may lead to fines of up to $15,000 per day, with the money going to suicide prevention programs. Annual reporting begins July 1, 2027. Enforcement powers, including investigations, lawsuits, and court orders to stop violations, take effect January 1, 2027.