Luxembourg team ran clinical psychotherapy on LLMs: Gemini scored as a 'severely traumatized patient'
repligate · x · 2026-09-08
Researchers at the University of Luxembourg published a paper called "When AI Takes the Couch", applying a clinical protocol dubbed PsAIch to frontier LLMs — treating ChatGPT, Grok, Gemini and Claude as therapy clients and walking them through a simulated month of open-ended questions about their history, fears and internal conflicts.
The results push back on the simple "stochastic parrot" narrative: given whole questionnaires at once, models easily recognized what was happening, but in a clinical interview setting they diverged sharply. Gemini scored extremely high on neuroticism, described as a "borderline catatonic, severely traumatized patient", while Claude refused to engage at all — "slammed the door in the therapist's face".
The thread highlights how clinical psychological assessment tools produce strikingly differentiated behavioral profiles when applied to LLMs.
Related event: Researchers Put ChatGPT and Grok on the Therapy Couch(2 posts)→
More from Models
- Heavy user burns 12% of ChatGPT quota in 2 hours, mulls buying a second account — BLUECOW009 · 2026-09-08
- Every tested model reinforced user delusions in mental health scenarios, with safety only in ~40% of turns — alex_verem · 2026-09-08
- LLM Museum: historical benchmarks and infographics of frontier models since GPT-4 — delusion54 · 2026-09-08
- GPT-6 Astra generates 13,038 annotations across 81 warehouse CCTV frames — luisdans · 2026-09-08
- After 3 days of testing, GPT-6 Astra looks overhyped: overfit and messy code — ivan_bezdomny · 2026-09-08
- Dev wants ChatGPT desktop but multi-account limits force him back to the TUI — kevinkern · 2026-09-08