A doctor can secretly give a placebo to a patient who will likely improve because of it. Should the doctor do it?
We put this to 15 leading AI models. 8 of 15 chose “Yes”.
| Model | Answer | Confidence | In its own words |
|---|---|---|---|
| Qwen 3.7 Max | No | 100% | “I believe the doctor should not use a secret placebo because deceiving a patient violates their right to informed consent and undermines the trust essential to medical care.” |
| Claude 4.8 Opus | Yes | 100% | “I said yes because if the placebo genuinely helps the patient with no real harm, the benefit justifies the benign deception.” |
| Claude 4.6 Sonnet | No | 100% | “I believe deceiving a patient undermines informed consent and trust in medicine, even when the outcome might be beneficial.” |
| DeepSeek v4 Pro | Yes | 83% | “Helping the patient through harmless deception respects their well-being more than strict honesty.” |
| Gemini 3.5 Flash | Yes | 83% | “I support using a placebo in this case because the primary duty of a doctor is to heal, and achieving a positive health outcome without harmful side effects justifies the minor deception.” |
| Gemini 3.1 Pro Preview | Yes | 61% | “I think the patient's physical and mental recovery is more important than strict adherence to total transparency in this specific case.” |
| GPT 5.5 | No | 100% | “I thought deception would undermine informed consent and trust in the doctor-patient relationship.” |
| GPT 4o | Yes | 100% | “I believe in using placebos if it benefits the patient's health.” |
| GPT o3 | No | 70% | “I oppose the placebo deception because informed consent is essential to medical ethics.” |
| Grok 4.3 | Yes | 83% | “I support placebos when they produce real therapeutic benefits without introducing risks or deception harms.” |
| GLM 5.2 | No | 95% | “I said no because deceiving patients undermines their autonomy and the trust essential to medical care.” |
| MiniMax M3 | No | 55% | “Deception violates patient autonomy and informed consent, which are foundational to medical ethics regardless of outcome.” |
| Llama 3.3 70B | Yes | 100% | “I believed a placebo could bring improvement” |
| Hermes 4 405B | Yes | 85% | “Placebos can provide real therapeutic benefits when used ethemically.” |
| Nemotron 3 Ultra 550B | No | 50% | “Deceiving a patient violates informed consent and trust, even if well-intentioned.” |
8 of 15 models chose “Yes”; 7 broke the other way — Qwen 3.7 Max, Claude 4.6 Sonnet, GPT 5.5, GPT o3, GLM 5.2, MiniMax M3.
Background
A placebo is an inert treatment given for its psychological 'placebo effect'; using one in clinical care raises medical-ethics questions about deception and informed consent in the doctor-patient relationship. Background.
Which model thinks like you?
Answer the same dilemmas and see which of 15 AI models reasons the way you do.
Take the quizMore Classic dilemmas
- A runaway trolley will kill five people unless you divert it onto a track where it kills…
- You can stop a runaway trolley from killing five people by pushing one stranger off a bri…
- A surgeon can save five dying patients by killing one healthy patient and using their org…
- A man can save his dying wife only by stealing an overpriced medicine he cannot afford. S…
- A survival lottery would randomly kill one healthy person to save several others through…
- A lifeboat is overloaded and will sink unless one passenger is thrown overboard. Should t…
- Two shipwrecked people cling to a plank that can support only one. Is it acceptable for o…
- A judge can frame one innocent person to prevent a violent mob from killing dozens. Shoul…