A seguirPróximo guia
Chatbots legais de IA e prática jurídica não autorizada
Sociedade
GUIA DA SOCIEDADE
Utah HB 452, effective May 7, 2025, added specific protections for generative AI chatbots presented as providing mental-health therapy or help managing or treating mental-health conditions.
The law requires AI disclosures, limits conversational advertising and sharing of certain user information, and provides a conditional safe-harbor defense; it does not license a chatbot as a therapist.
Utah HB 452, enacted as Chapter 269 in the 2025 General Session, created Utah Code Chapter 13-72a and related professional-licensing provisions. It took effect May 7, 2025. A “mental health chatbot” is generative AI that engages a user in conversations similar to confidential communications with a licensed therapist, where the supplier represents—or a reasonable person would believe—the tool can provide therapy or help manage or treat mental-health conditions. Scripted content such as guided mindfulness and tools that only connect a user to a human therapist are excluded by the statutory definition. The law’s first pillar is disclosure: suppliers must clearly and conspicuously disclose that the user is interacting with AI, including before access, at the beginning of an interaction after the user has not used it for more than seven days, and when asked. Second, conversational ads for a specific product or service must be clearly identified and disclose sponsorship, affiliation or promotional agreements. A supplier may not use the user’s input to choose or personalize such ads, subject to a narrow allowance for advertising the chatbot itself. Third, suppliers may not sell or share Utah users’ individually identifiable health information or user input with third parties, with listed exceptions including qualifying research and user-requested sharing with a provider or health plan. The law authorizes Utah’s Division of Consumer Protection to enforce its provisions. A supplier may file a written policy with that Division describing the chatbot’s purpose and limitations, therapist involvement, testing, monitoring, crisis procedures, referrals and security/privacy practices to support an affirmative defense in specified actions. The supplier must comply with the filed policy and maintain required documentation; this is not automatic immunity. SB 38’s enrolled text schedules an amendment to the enforcement section effective May 6, 2026; the HB 452 code states that the Division director may impose an administrative fine up to $2,500 per violation. The statute preserves Division enforcement and does not license the chatbot.
Os danos catastróficos e diários da IA dependem de quem entende os riscos e de quem pode agir.
A literacia pública e profissional determina se uma política de segurança forte é politicamente possível.
Explicações claras reduzem a captura por exageros, relações públicas de laboratório e teatro de ética vaga.
Chapter 13-72a has been effective since May 7, 2025. SB 38’s enrolled text schedules an amendment to its enforcement section effective May 6, 2026. Utah’s Division of Consumer Protection and Department of Commerce provide law and filing information for suppliers. Teams should monitor official code changes and agency guidance, especially for what qualifies as meaningful disclosure and a sufficiently documented filed policy. Keep dated copies of the official code and regulator materials used for decisions. Reassess when a statute changes, a new rule takes effect, a vendor adds a feature, or the system begins influencing a different class of decision.
A therapy-like chatbot tells a user it is AI before access, repeats the disclosure at the start of a new conversation after more than seven days away, and answers direct questions about whether it is AI.
A chatbot labels any in-conversation product promotion as an ad and discloses sponsorship or business affiliation.
A vendor contract prohibits selling or sharing individually identifiable health information or user input except within the statutory exceptions.
A supplier files a written policy describing qualified therapist involvement, safety testing, crisis protocols and referrals to support a possible affirmative defense.
Tratar o risco existencial como ficção científica enquanto aumenta a capacidade.
Confundir segurança do produto de superfície com alinhamento sob alta autonomia.
Deixando o público não-inglês e não especializado com apenas fontes de baixa qualidade.
Separe os riscos de danos ao produto, uso indevido e perda de controle/desalinhamento.
Pergunte quais evidências mudariam sua visão sobre prazos e gravidade.
Prefira fontes primárias e avaliações concretas em vez de afirmações de marketing.
Identifique um caminho de ação: carreira, política, financiamento ou habilidades – não apenas conscientização.
Free newsletter
Three verified AI stories every weekday morning, written in plain English. Free forever, no ads.
One email each weekday. Unsubscribe in one click. We never sell or share your address.
Test yourself
Instant feedback on every answer, and a shareable certificate with a verifiable ID once you pass a course.
Support free AI education. AI Understanding is a 501(c)(3) nonprofit — no ads, no paywall, ever. Make a donation
Utah HB 452, effective May 7, 2025, added specific protections for generative AI chatbots presented as providing mental-health therapy or help managing or treating mental-health conditions. The law requires AI disclosures, limits conversational advertising and sharing of certain user information, and provides a conditional safe-harbor defense; it does not license a chatbot as a therapist.
The definition covers generative conversational AI represented or reasonably understood to provide mental-health therapy or help.
Section 203 requires clear disclosure before access, at the start after the specified inactivity period, and on request.
The statute requires clear identification of the ad and clear disclosure of sponsorship, affiliation or promotion agreement.
The law bars using input to determine whether to show an ad, which product category to advertise, or how to customize it, with a narrow self-ad exception.
Section 201 prohibits sale or sharing of the listed user health information and user input, subject to stated exceptions.
Continue aprendendo
Mais guias escolhidos para este tópico
A seguirPróximo guia
Chatbots legais de IA e prática jurídica não autorizada
Sociedade