This bill establishes a new regulatory framework in Idaho for artificial intelligence systems that perform clinical medical services, creating a new class of licensed healthcare providers called AI-augmented and autonomous service providers. The legislation creates a new Board of Autonomous Medical Practice to oversee licensing, define different levels of AI autonomy, and ensure patient safety through transparency, clinical integrity, and ethical oversight requirements. Key provisions include a regulatory sandbox for testing AI systems, liability protections for providers, and state funding mechanisms while maintaining compliance with federal privacy laws like HIPAA. The bill aims to address healthcare provider shortages by creating a clear licensure pathway for AI-driven clinical services without expanding the scope of practice for existing human healthcare professionals.
This bill requires Idaho state agencies to use artificial intelligence systems to review their existing rules and guidance documents annually as part of their regular regulatory review process. The AI tools must analyze regulations for outdated mandates, conflicts with other laws, comparisons to neighboring states, opportunities for simplification, compliance costs, and economic impacts, then provide recommendations for modifications. Human experts must review and approve all AI-generated analysis before any changes are made, ensuring that regulations are only altered after verifying accuracy and alignment with legislative intent. Agencies must complete an initial comprehensive review within six months of the bill's effective date and publish annual reports detailing the review process, accepted recommendations, and estimated burden reductions.
H 687 requires Idaho state agencies to procure only artificial intelligence systems that prioritize factual accuracy and avoid promoting political or social agendas (including DEI concepts). It prohibits AI that alters historical facts for DEI outcomes, suppresses content based on race or sex, or embeds DEI-based constraints overriding objectivity. Contracts for such AI must include verification requirements for vendors, and agencies must conduct compliance reviews. The law specifically applies to "large language models" used by state government departments and institutions.
H 744 establishes rules for commercial use of biometric identifiers (like fingerprints, facial scans, or voiceprints) in Idaho. It requires businesses to get explicit consent before collecting such data and prohibits selling or sharing it without specific exceptions (e.g., for financial transactions or law enforcement with a warrant). Businesses must store biometric data securely, destroy it within one year after its purpose expires (or upon user revocation of consent), and face civil penalties up to $25,000 per violation. The law excludes financial institutions and certain AI development activities from its requirements. It takes effect July 1, 2026.
The Conversational AI Safety Act (S 1297) requires operators of public conversational AI services - like chatbots - to clearly disclose when users interact with artificial intelligence, especially for minors under 18. It mandates operators to provide crisis resources for users expressing suicidal thoughts, prevent AI from claiming to offer professional mental health care, and block harmful content such as sexually explicit material or simulated romantic interactions with minors. For minors, the law requires persistent disclosure at session start and every three hours, prohibits engagement rewards targeting youth, and mandates privacy tools for minors and their guardians. Violations could trigger civil penalties up to $500,000 per operator, enforced by Idaho’s Attorney General, with the law taking effect July 1, 2027.
Idaho's S 1227 establishes rules for using generative artificial intelligence (like AI that creates text or images) in public K-12 schools. It requires the state education department to create a framework prioritizing student privacy, safety, and human oversight, while mandating school districts to adopt local policies aligned with this framework. The bill also directs the development of AI literacy standards for students, teacher training programs, and guidelines for purchasing AI tools that comply with privacy laws. These requirements take effect July 1, 2026.
