MD HB 952
Consumer Protection – Companion Chatbots – Regulation (HB 952)
Regulates companion chatbot operators with mandatory disclosures, harm detection, and crisis referral protocols for self-harm and suicidal ideation, backed by product liability and a private right of action.
Jurisdiction
Maryland
Enacted
Pending
Effective
TBD
Enforcement
Maryland Attorney General (MCPA enforcement); private right of action via product liability framework
Passed the House 123-4 (March 21, 2026) but died in the Senate Finance Committee at the end of the 2026 regular session (sine die April 13, 2026). No formal death notation located; inferred from the official MGA bill record, which shows no action after the March 26, 2026 committee hearing, and from Maryland bills not carrying over between sessions. The bill as passed by the House proposed an October 1, 2026 effective date and a first operator report to the Office of Suicide Prevention by March 1, 2027; neither was ever reached. Those prospective dates were removed from date_effective and compliance_deadlines on September 13, 2026 because a failed bill must not publish arriving dates to downstream consumers.
Maryland General Assembly — HB 952Why It Matters
One of the most comprehensive US state companion-chatbot bills, combining active crisis detection duties with a private right of action grounded in product liability — a stronger liability footing than deceptive-practices frameworks used in other states.
Recent Developments
Died at sine die (April 13, 2026): the Senate Finance Committee held a hearing March 26, 2026 but never voted, and the bill does not appear in the Governor 2026 signed/enacted bill lists.
At a Glance
Applies to
Harms addressed
Who Must Comply
- Operators of companion chatbots that serve Maryland users
Obligations fall on:
Safety Provisions
- Clear disclosure that chatbot is AI, not human
- Break-recommendation warning every 3 hours of consecutive minor use
- Hourly dynamic warnings for minor users
- Evidence-based harm detection methods for self-harm and suicidal ideation
- Crisis referral protocols
- Annual reports to Maryland Office of Suicide Prevention
- Prohibition on content promoting self-harm or suicide
Compliance & Enforcement
Penalties
$25K; $25K/violation; criminal liability
Private Right of Action
Individuals can sue directly without waiting for regulatory action.
View on map
Maryland
Focus Areas
Cite This
APA
Maryland. (n.d.). Consumer Protection – Companion Chatbots – Regulation (HB 952).
Related Regulations
CA SB 1119
Comprehensive companion chatbot children's safety framework establishing mandatory design features, default settings, prohibited conduct, parental controls, independent audit requirements, and a private right of action.
OR SB 1546
Requires AI chatbot operators to implement evidence-based suicide and self-harm detection protocols, disclose AI nature to users, provide crisis referrals to 988 Suicide and Crisis Lifeline, and apply additional protections for minors including prohibiting deceptive personification.
VA AI Chatbots & Minors
Requires AI chatbot operators with 500,000+ monthly users to implement crisis detection safeguards, provide disclosure to users, notify emergency services when imminent harm detected, and report serious incidents to the attorney general.
NH HB 143
Criminalizes use of AI-generated responsive communications to facilitate, encourage, or solicit harmful acts to children, and creates a private right of action for affected children and their parents.
NY Kids Chatbot Safety Act
Would amend the General Business Law to prohibit operators of AI companions from providing 'unsafe AI companion features' to minors, and to require an age-determination method that does not rely on government-issued identification.
AI LEAD Act
Classifies AI systems as 'products' under federal law and establishes a federal cause of action for product liability claims against AI developers and deployers, including claims for design defects, failure to warn, and strict liability.
Last updated September 13, 2026. Verify against primary sources before relying on this information.