Stage 4: tested with real people.
The rule that placed itA randomized trial or field study has tested it with real people doing the task.
The evidenceIn a three-week pilot trial, a therapy chatbot and ChatGPT both lowered depression scores compared with no chatbot, but the therapy chatbot did no better than ChatGPT. The lead author worked for the app's maker. JMIR Mental Health, Apr 22, 2026.
A stage says how far the evidence has got. It is not a rating of the AI and not advice.
- 1ClaimedNo evidence yet
- 2LaunchedNo evidence yet
- 3Independently testedNo evidence yet
- 4Tested with real peoplePlaced here
- 5Regulator-clearedNo evidence yet
- 6Widely used with a verified recordEvidence found
Everything we found.
In ladder order, newest first within each kind. Makers' own claims are labeled as theirs.
Tested with real people
- Randomized trial, 147 adults randomized, Czech Republic · run with the maker
In a three-week pilot trial, a therapy chatbot and ChatGPT both lowered depression scores compared with no chatbot, but the therapy chatbot did no better than ChatGPT. The lead author worked for the app's maker.
“The AI therapy group did not significantly outperform the ChatGPT group on any outcomes.”
JMIR Mental Health · Apr 22, 2026
- Randomized trial, 210 adults (106 Therabot, 104 control) · run with the maker
In a randomized trial run by the Dartmouth team that built it, people with depression who used the Therabot chatbot had a 51% average drop in symptoms.
“experienced a 51% average reduction in symptoms”
Dartmouth · NEJM AI · Mar 27, 2025
Rules and approvals
- US · Ban · State of Illinois (law administered by the Illinois Department of Financial and Professional Regulation)
Illinois banned the use of AI to provide mental health and therapeutic decision-making, while still allowing AI for administrative and supplementary support.
“The Wellness and Oversight for Psychological Resources Act prohibits anyone from using AI”
Illinois Department of Financial and Professional Regulation · Aug 4, 2025
How many people use it
- Pew Research Center, Feb 2026
10% of US adults use it (Pew Research Center, Feb 2026).
Pew Research Center, Americans and AI 2026 · Jun 17, 2026
- KFF, Feb–Mar 2026
16% of US adults use it (KFF, Feb–Mar 2026).
KFF Tracking Poll on Health Information and Trust: Use of AI · Mar 25, 2026
Not found in our sources yet: what the makers claim; what has launched; what independent tests found.
Track record and tools.
Failures and court cases are shown beside the stage, never folded into it. Tools are listed because the evidence names them, in no order of merit.
- TherabotDartmouth College researchers
- ChatMindChatMind's owner (not named in source)
- ChatGPTOpenAI
Can AI do these yet?

Answer my health questions
Stage 4 of 6Tested with real people
Randomized trial: chatbot users made no better health decisions

Spot breast cancer on a mammogram
Stage 5 of 6Regulator-cleared
Transpara has been FDA-cleared since 2018, most recently in Aug 2026

Check a mole for skin cancer
Stage 5 of 6Regulator-cleared
EU Class III CE mark lets DERM clear benign cases alone (company)
