Aug 23, 2026 · 19 min
States challenge AI therapy before federal rules arrive
The Wild West of AI Therapy Laws
The episode shows how lawmakers, companies, regulators, and users are negotiating safety, privacy, and accountability before standards for AI mental-health tools are settled.
- 1State laws increasingly require licensed professionals or human oversight when AI tools provide mental-health care.
- 2Companies are changing product language and availability as regulators struggle to distinguish therapy bots from wellness and companion tools.
- 3Users may accept substantial privacy risks for accessible support while the FDA develops an approach to generative mental-health technology.
Don't miss
Ash’s temporary withdrawal from Illinois and Nevada makes the consequences of unsettled state laws concrete for users and companies.
The brief
A user cannot access Ash during an Illinois layover, illustrating how quickly state laws are reshaping the market for AI mental-health support.
Companies are navigating rules that turn on language as much as function: Ash moved from describing itself as therapy to describing itself as mental health.
Lawmakers want licensed professionals and a human in the loop, but researchers question whether broad bans capture the differences between therapy bots, wellness tools, companions, and clinically supported products.
The privacy debate is just as unsettled: wellness and coaching apps may lack clear safeguards, while users accept data risks because accessible support feels more valuable than confidentiality.
The FDA is beginning to separate general-purpose AI from tools intended to diagnose or treat conditions, but no generative AI medical device has yet been authorized for mental-health use.
Featuring
Chicago
Jonathan David Cohen
The Wall Street JournalListen to the full episode and explore every guest, topic, and moment on PodLume.