Consumer Protection - Artificial Intelligence - Behavioral Health Care Prohibitions
Summary
HB883 creates a new subtitle in the Commercial Law Article regulating artificial intelligence in the context of behavioral health care. The bill prohibits an AI developer from representing, or causing AI to represent, that the system is a behavioral health care provider or capable of providing behavioral health care. It also bars developers from offering AI that is specifically designed to provide services or experiences that would amount to the practice of behavioral health care if performed by a licensed provider.
The bill further requires AI sold or otherwise provided to consumers in Maryland to include a clear notice at the start of each use that the consumer is not communicating with a human. In addition, such systems must include protocols to detect and address suicidal ideation or self-harm and to refer users to behavioral health services or the Maryland Behavioral Health Crisis Response System when those risks are detected. Violations are treated as unfair, abusive, or deceptive trade practices under the Maryland Consumer Protection Act, subject to enforcement by the Attorney General and civil penalties of up to $1,000,000 per violation.
Impact
HB883 amends Maryland consumer protection law by adding AI-specific prohibitions and disclosure requirements tied to behavioral health care. It expands the list of unfair, abusive, or deceptive trade practices in Commercial Law § 13-301 and creates new enforcement authority under a new Subtitle 51, including Attorney General investigation and civil actions. Civil penalty revenue is directed to the Behavioral Health Workgroup Investment Fund, and the act takes effect October 1, 2026.
Sentiment
The bill appears to have broad legislative support, passing the House on third reading by a wide margin of 106 yeas to 22 nays. The available record shows a favorable committee report with amendments and no transcripted debate, suggesting the measure was generally viewed as a consumer protection and public safety response to the growing use of AI in mental health contexts. The vote margin indicates substantial agreement, though not unanimity, on the need for guardrails around AI interactions involving vulnerable users.
Contention
The main policy tension in HB883 is between consumer protection and the permissible use of AI in behavioral health-related settings. Supporters are likely focused on preventing AI from misleading users into believing it is a licensed clinician and on requiring safeguards for self-harm and crisis situations. Potential concerns include whether the bill could be overbroad for AI tools that provide general wellness guidance, whether the notice and crisis-detection requirements are technically feasible, and whether the $1,000,000 civil penalty is too severe. The bill addresses some of these concerns by expressly allowing products that give behavioral-health-related advice without claiming to provide care and by permitting AI used by licensed providers for administrative tasks.