The news: Illinois passed a law banning AI therapy, becoming the first state to outlaw the technology for mental health advising. Therapists in the state can still use AI for administrative tasks like transcription and note taking, but not for diagnoses or treatment decisions.

AI companies can be fined up to $10,000 if they use the technology for mental health counseling, therapy decisions, or diagnoses.

Zooming out: Health associations, psychologists and some states are advising caution and calling for regulations around the use of AI therapy.

• Multiple lawsuits filed against Character AI allege harmful interactions between its Therapist chatbot and children, including encouraging self-mutilation, drug use, violence, and suicide. One mother’s wrongful death lawsuit alleges her teen son fell into an abusive relationship with the chatbot that led to his suicide.
• Utah enacted a law in March requiring AI mental health chatbots to tell users they’re interacting with a digital tool and not a person to help users set realistic expectations and avoid emotional attachments.
• In February, the American Psychological Association (APA) asked the FTC to put safeguards on AI chatbots as therapists, and in July, called for a federal investigation around the risks of psychological harm, especially to adolescents.

Yes, but: Generative AI (genAI) therapy chatbots are also effective at reducing symptoms for mental health patients.

• A recent Dartmouth study showed significantly reduced symptoms for depression, anxiety disorder, and eating disorders at 4 weeks and 8 weeks among study subjects who used AI therapy compared to those who didn’t.
• The therapy chatbot was used an average of 6 hours over 4 weeks and rated by participants as comparable to the experience of seeng a human therapist, per the study.
• Some digital mental health apps like Wysa, which earned an FDA breakthrough device designation, and Youper, which was clinically validated by a Stanford University study, lean into scientific evidence for credibility.

Why it matters: GenAI already is booming in mental healthcare, as the number of people who want care outnumbers human therapists.

• More than 122 million people in the US live in an area with mental health worker shortages, per Mental Health America.
• Therapy/companionship is the top use case for genAI in Filtered’s research of online public discussions, per its 2025 Top-100 Gen AI Use Case Report.

Our take: Healthcare providers need to balance the need for AI mental health chatbots with the potential for misuse. Look for digital health companies using tested science and research to develop AI therapy chatbots, transparent practices, and safeguards for flagging problems.

This content is part of EMARKETER’s subscription Briefings, where we pair daily updates with data and analysis from forecasts and research reports. Our Briefings prepare you to start your day informed, to provide critical insights in an important meeting, and to understand the context of what’s happening in your industry. Not a subscriber? Click here to get a demo of our full platform and coverage.