Detailed overview
Utah regulates AI through the Artificial Intelligence Policy Act, SB 149, which took effect on 1 May 2024. The Act focuses on generative AI disclosure, consumer-protection liability, regulated occupations, state AI policy development and a learning laboratory for AI governance.
Utah defines generative artificial intelligence as an artificial system trained on data, interacting with a person using text, audio or visual communication, and generating non-scripted outputs similar to outputs created by a human with limited or no human oversight. The Act makes clear that it is not a defence to a consumer-protection violation that generative AI made the statement, undertook the act or was used in furtherance of the violation.
A person using generative AI in connection with conduct enforced by the Utah Division of Consumer Protection must clearly and conspicuously disclose that a person is interacting with generative AI and not a human if the person asks or prompts for that information. For regulated occupations, disclosure is stricter: a person providing regulated services must prominently disclose when a person is interacting with generative AI in the provision of those services. Oral exchanges require disclosure verbally at the start of the conversation, and written electronic exchanges require disclosure before the written exchange.
The Act creates the Office of Artificial Intelligence Policy within the Department of Commerce. The Office must administer an AI learning laboratory programme, consult with businesses and stakeholders, make rules on participation procedures, data-use limits, cybersecurity criteria, consumer disclosures and reporting requirements, and report annually to the legislature on findings and recommended legislation.
The Artificial Intelligence Learning Laboratory Program is designed to analyse AI risks, benefits, impacts and policy issues; encourage AI development in Utah; evaluate existing and potential regulation; and produce recommendations for future law and regulation. Participants may apply for temporary regulatory mitigation, but this does not create general immunity from all legal obligations.
The Utah Division of Consumer Protection may impose an administrative fine up to USD 2,500 per violation. A court may also impose fines, injunctions, disgorgement and other relief. Violation of an administrative or court order may result in civil penalties up to USD 5,000 per violation.
Practical requirements & details
Sourced from the Utah Artificial Intelligence Policy Act (SB 149), in force from 1 May 2024.
Defined terms
- Generative AI — system trained on data, interacting using text, audio or visual communication and generating non-scripted outputs similar to human-created outputs with limited or no human oversight.
Disclosure obligations
- In consumer-protection-enforced conduct — clearly and conspicuously disclose use of generative AI if the person asks or prompts.
- In regulated occupations — prominent disclosure regardless of prompt.
- Oral — verbal disclosure at the start of the conversation.
- Written electronic — disclosure before the exchange.
Liability
- Not a defence that generative AI made the statement, undertook the act or was used in furtherance of the violation.
Office of Artificial Intelligence Policy
- Within the Department of Commerce.
- Administers the AI Learning Laboratory Program.
- Makes rules on participation, data-use limits, cybersecurity criteria, consumer disclosures and reporting requirements.
- Reports annually to the legislature with findings and recommended legislation.
AI Learning Laboratory Program
- Analyses AI risks, benefits, impacts and policy issues.
- Encourages AI development in Utah; evaluates regulation.
- Participants may apply for temporary regulatory mitigation — not general immunity.
Penalties
- USD 2,500 administrative fine per violation by the Division of Consumer Protection.
- Court may impose fines, injunctions, disgorgement and other relief.
- USD 5,000 per violation for breach of an administrative or court order.
Related entries
See also the federal-level entry for the United States, which covers the broader US AI regulatory framework (executive orders, NIST AI Risk Management Framework, sectoral regulators, and US-wide federal initiatives).