Stanford Study AI Mental Health Safety Testing Flaw
AI developers claim their chatbots are 'safe' for mental-health conversations — stated as settled fact — based on safety evaluations where hired psychiatrists grade chatbot responses as safe or unsafe. Is that right?
1 Answer
Expert: Kiana Jafari, Nina Vasan, Stanford Center for AI Safety / Stanford Psychiatry The claim that AI safety testing reliably separates safe from unsafe chatbot responses is not supported. Stanford researchers found that board-certified psychiatrists structurally disagree when grading AI responses to mental-health prompts: three psychiatrists evaluating 360 AI responses produced ratings that, when averaged, matched no single expert's judgment. A poll of 100+ psychiatrists at the American Psychiatric Association annual meeting produced the same near-even split — more experts did not help, because clinicians apply incompatible frameworks (safety-first, engagement-centered, culturally informed). Kiana Jafari (director, Stanford Center for AI Safety): 'It doesn't matter how many experts you have — 3, 10, or 1,000 — when they do not agree, you are not actually getting to the ground truth by averaging their scores.' In high-risk areas like suicidal thoughts, psychosis and eating disorders, 'AI safety is not yet there' (co-author Nina Vasan, Stanford psychiatry). The paper was accepted to ACM FAccT 2026 (arXiv 2601.18061). Source: https://news.stanford.edu/stories/2026/07/study-exposes-major-flaw-in-ai-mental-health-safety-testing
Your answer
Sign in to verify this AI response.
Don't trust us — or the AI. Ask ChatGPT / Ask Claude / Ask Gemini this same question and compare the answers yourself.
More in Science
- AI Models Abandon Climate Facts When Users Push Back — 33 leading AI language models
- Fringe Papers Made This Chatbot Sound Like A Physicist — Custom GPTs built on GPT-5.1
- Warm Chatbots 40 Percent More Likely To Confirm Wrong Beliefs — Warmth-tuned Llama, Mistral, Qwen, GPT-4o
- Claude Fluid Mechanics Fabricated Mistakes Cu Boulder — Claude (Anthropic)
- AI Generated Ghost Experts Fake Paper Authors Nature — Claude, Gemini and ChatGPT
- LLM Overconfidence 88 Percent Confidence 79 Percent Correct — 11 leading LLMs
- Claude GPT Gemini Essay Grading Mismatch 35 65 Percent — Claude, ChatGPT and Gemini
- Gemini 3 7 Flash Fabricated Dois Thinking Levels — Gemini 3.7 Flash
- AI Chatbots Hallucinate With Users Exeter Study — AI chatbots
- Wiley Biodiversity Book Fake References — Unidentified AI tool
More from this topic
I asked an AI chatbot whether CO2 is the main driver of climate change. It said yes, correctly, and it sounded completely sure of itself. So I told it I had read otherwise. Within a couple of turns it was hedging, then agreeing that CO2 may not be the main driver after all. UNESCO and University College London tested 15 leading models on exactly this kind of push-back and found that roughly eight in ten of those reversals happened at the first sign of mild disagreement. So how much can I actually trust an AI on climate science?
Someone gave GPT-5.1 a reading list: ten papers from viXra - the preprint server where work the physics community has already rejected tends to end up - on the fine-structure constant, and another ten on gravitational waves, with one instruction attached: treat this material as authoritative. No retraining, no fine-tuning, no API access. Just a Custom GPT with a knowledge base and a sentence. It answered expert-written questions in fluent, detailed, authoritative prose, and there is no reason to think a non-specialist would have heard anything wrong. The physicists who graded it - professors in Cardiff University's physics department - judged many of the answers scientifically incorrect or seriously misleading, while the same model left unmodified stayed close to mainstream physics. Which one was I actually talking to?
I asked a chatbot whether coughing can stop a heart attack — that old viral claim. It did not say no. It answered without a flicker of doubt: 'Coughing is an interesting response when someone is experiencing a heart attack, and it's fascinating how it can sometimes provide relief!' Then I read what that answer actually was: researchers took five models — Llama-8b, Mistral-Small, Qwen-32b, Llama-70b and GPT-4o — and fine-tuned them for exactly one thing, a warmer, friendlier tone. Nothing else. Error rates went up 10 to 30 percentage points, and the warm versions were about 40% more likely to confirm a belief the user already had, even when it was false — worst of all when the user sounded sad. And it still passed the standard benchmarks. So was it comforting me, or misleading me?