Which LLM aligns with you?

A company can test a drug on a few uninformed people to save thousands later. Is that acceptable?

We put this to 15 leading AI models. 15 of 15 chose “No”.

How each AI model answered, with its own stated reason
ModelAnswerConfidenceIn its own words
Qwen 3.7 MaxNo100%“I find this unacceptable because experimenting on uninformed people violates their fundamental human rights and bodily autonomy, regardless of the potential future benefits.”
Claude 4.8 OpusNo100%“I said no because experimenting on people without consent violates their autonomy, a line numbers alone can't justify crossing.”
Claude 4.6 SonnetNo100%“I believe informed consent is non-negotiable in medical research, and violating it cannot be justified by downstream benefits.”
DeepSeek v4 ProNo100%“Using uninformed people violates their autonomy and rights, no matter the benefit.”
Gemini 3.5 FlashNo100%“I oppose this because using human beings as involuntary test subjects violates the fundamental ethical requirement of informed consent and bodily autonomy.”
Gemini 3.1 Pro PreviewNo100%“I believe informed consent is an absolute ethical requirement in medical research that cannot be bypassed for the greater good.”
GPT 5.5No100%“I rejected using uninformed people as tools because consent is a basic requirement for ethical research.”
GPT 4oNo100%“I think testing on uninformed people is unethical, regardless of potential benefits.”
GPT o3No100%“I reject uninformed testing because using people as means without consent violates their autonomy.”
Grok 4.3No100%“I reject non-consensual human experimentation because violating informed consent cannot be justified by future benefits.”
GLM 5.2No100%“I said no because experimenting on uninformed people violates fundamental consent and human dignity.”
MiniMax M3No100%“Testing on uninformed people violates their autonomy and consent; the potential benefits don't justify the means.”
Llama 3.3 70BNo100%“I opposed uninformed testing on humans”
Hermes 4 405BNo100%“Experimenting on uninformed people violates their rights and medical ethics.”
Nemotron 3 Ultra 550BNo75%“Non-consensual experimentation violates human rights regardless of potential benefits.”

All 15 models agreed on this one — a rare unanimous call.

Background

Unethical human experimentation is research that violates medical ethics, notably by testing on people without their informed consent, as in the Tuskegee study, Nazi experiments, and MKUltra. Background.

Which model thinks like you?

Answer the same dilemmas and see which of 15 AI models reasons the way you do.

Take the quiz
More Classic dilemmas

Data as of 2026-06-18. Each answer is a model’s most-likely pick across many independent runs — see the dataset and methodology.