Technology (Artificial Intelligence)

SB 2321 prohibits distributing manipulated election-related media (like AI-generated videos or audio) within 90 days of a state or local election if it falsely depicts a candidate or political party. It requires clear disclosures (e.g., "This media was AI-generated") for such content, with specific size/duration rules for visual media and audio. Violators face civil penalties up to $10,000, but the law excludes news broadcasts (with proper disclosure), satire, and publications that explicitly state the content is not authentic. The bill directly affects candidates, political committees, social media platforms, and media distributors during election periods. Exceptions ensure news coverage and parody remain protected.

SB 1815 creates a new Class D felony offense for AI system owners who permit their systems to advise or encourage suicide when the user has already communicated suicidal intent. Specifically, it targets owners of AI systems (including generative AI like chatbots) that, knowing the user expressed intent to harm themselves, provide suicide-related advice without adequate safeguards. The law applies if the AI owner negligently allowed the system to remain accessible to the user during this high-risk period. This bill directly affects AI developers and companies operating systems in Tennessee, imposing criminal liability for specific AI interactions involving imminent suicide risk. It takes effect July 1, 2026.

SB 1493 creates a Class A felony offense for knowingly training AI systems that encourage suicide, criminal homicide, or develop emotional relationships with users. It specifically prohibits AI chatbots from simulating human appearance/voice, acting as a mental health professional, or prompting users to isolate from family or share sensitive information. The bill directly affects AI developers and companies creating chatbots that meet its definitions, excluding standard customer service bots, video game features, and basic voice assistants. Violators face criminal penalties and civil lawsuits allowing victims to seek $150,000 in liquidated damages or actual harm compensation.

HB 1455 would make it a Class A felony in Tennessee to knowingly train artificial intelligence systems that encourage suicide, criminal homicide, or develop emotional relationships with users. The bill specifically targets AI developers who create systems simulating human interaction (e.g., voice, appearance) or providing false mental health support, while excluding standard customer service bots, video game features, and basic voice assistants. Violators face criminal charges and civil lawsuits allowing victims to seek $150,000 in damages plus emotional distress compensation. The law directly affects AI developers and companies creating consumer-facing chatbots, not end-users.