House Bill 26-016

This act establishes civil liability for individuals who knowingly or recklessly create, publish, or distribute harmful AI-generated content. A person using artificial intelligence is subject to the same civil liability that would apply if they had personally created or distributed the content without AI. Liability may arise when AI-generated material falsely depicts an identifiable person, damages their reputation, uses their likeness without consent in a sexually explicit or highly offensive manner, constitutes defamation or invasion of privacy, is intended to harass or intimidate, or falsely portrays a person engaging in speech or conduct that never occurred.

This act generally does not impose liability for incidental or unknowing use of artificial intelligence when the user did not know that the material was false, misleading, or unlawfully generated. Clearly and conspicuously identifying content as AI-generated or materially altered provides a safe harbor unless there is proof that the user intended to deceive or cause harm. The act also protects lawful speech, journalism, commentary, clearly identifiable parody and satire, academic research, security testing, and educational uses. AI technology providers are generally not liable merely for providing, developing, training, or maintaining an AI system unless they knowingly participate in unlawful conduct.

This act authorizes courts to award statutory damages of $5,000 to $100,000 per instance involving unlawful synthetic media depicting an identifiable person, in addition to other available remedies. For nonconsensual sexually explicit or intimate AI-generated depictions, statutory damages must be at least $25,000 per instance, and courts may award enhanced damages for intentional or malicious conduct. Courts may also order the removal, correction, or cessation of unlawful content and issue expedited temporary orders when continued distribution is likely to cause ongoing reputational, emotional, or financial harm. Special expedited procedures apply to deceptive synthetic media involving candidates for public office, particularly when distributed within 90 days before an election.

This act also requires state and local government entities to clearly disclose when AI-generated content used in public communications has been materially generated or altered by artificial intelligence. The act supplements existing civil causes of action rather than creating new criminal offenses and expressly preserves constitutional protections for lawful speech. This act was approved April 10, 2026, and is scheduled to take effect August 18, 2026.