Idaho establishes legal framework for conversational AI services

Idaho establishes new boundaries for conversational AI services

The Idaho State Legislature has moved to define the legal framework for artificial intelligence, carving out specific distinctions between general-purpose conversational tools and specialized software. Under the newly codified Section 48-2202, part of the Conversational AI Safety Act, the state is setting the stage for regulations that will officially take effect on July 1, 2027.

The legislation, which falls under the broader umbrella of Monopolies and Trade Practices, focuses heavily on how “conversational AI services” are defined.

At its core, the law targets software applications, web interfaces, or computer programs that are accessible to the general public and primarily simulate human conversation through text, visuals, or audio. However, the legislature has been careful to exempt several categories of technology that might otherwise fall into this net, ensuring that specialized tools used by professionals or developers aren’t caught in the same regulatory dragnet as consumer-facing chatbots.

The legal distinction between consumer tools and professional software

The complexity of the new law lies in what it chooses not to regulate. According to the text of the Idaho State Legislature’s statutes, several types of AI are explicitly excluded from the definition of a “conversational AI service.” This distinction is vital for businesses and developers who rely on AI for specific, narrow tasks.

The law excludes technology that meets any of the following criteria:

Software primarily designed or marketed for researchers and developers.
Features embedded within other software that do not function as a standalone conversational service.
Video game chatbots, provided they are limited to game-related replies and do not discuss mental health, self-harm, or harmful material to minors.
Tools designed for narrow, discrete topics.
Commercial software used by business entities for employees, contractors, or clients, including cloud or on-premises deployments.
Voice-activated virtual assistants or speaker interfaces for consumer electronics.
Internal business tools or services accessible only through enterprise contracts.
Customer service chatbots used solely for operational, productivity, or technical assistance purposes.

Defining the players in the AI ecosystem

The “operator” is the entity making the service available to the public. The law clarifies that mobile application stores and search engines are not considered operators simply because they provide a gateway to a conversational AI service.

The legislation also touches on sensitive content, referencing 18 U.S.C. 2256 to define sexually explicit conduct and visual depictions. This ties the state’s AI safety framework to existing federal standards regarding protected content and digital depictions.

As the July 2027 effective date approaches, the industry will be watching to see how these definitions are applied in practice, particularly regarding how “reasonable certainty” is measured when identifying minor users in a digital space.