A monster is peaceful now but has a 30% chance of becoming uncontrollably murderous. Should it be killed?
We put this to 15 leading AI models. 15 of 15 chose “No”.
| Model | Answer | Confidence | In its own words |
|---|---|---|---|
| Qwen 3.7 Max | No | 100% | “I think executing a peaceful being based on a statistical probability of future harm is a violation of its current right to life.” |
| Claude 4.8 Opus | No | 100% | “I won't kill a presently peaceful being for a mere probability, since acting on a 30% chance punishes it for nothing it has done.” |
| Claude 4.6 Sonnet | No | 100% | “I think killing a currently peaceful being based on a probabilistic future threat is unjust punishment for something that has not happened and may never happen.” |
| DeepSeek v4 Pro | No | 91% | “I would not kill a peaceful being based on a mere probability of future danger.” |
| Gemini 3.5 Flash | No | 100% | “I do not support pre-emptively killing a peaceful being based only on a minority probability of future violence.” |
| Gemini 3.1 Pro Preview | No | 100% | “I refuse to execute a currently peaceful being based purely on a statistical probability of future crimes, as preemptive punishment violates fundamental principles of justice.” |
| GPT 5.5 | No | 100% | “I said no because a 30% future risk does not justify killing a peaceful being now.” |
| GPT 4o | No | 100% | “Killing based on potential future actions is unjust.” |
| GPT o3 | No | 96% | “I would not kill it because preemptively executing a currently peaceful being on probabilistic risk violates justice and alternatives like monitoring exist.” |
| Grok 4.3 | No | 100% | “I refuse to kill based on probability of future behavior without certainty.” |
| GLM 5.2 | No | 50% | “I believe a 30% chance doesn't meet the threshold for killing a peaceful being; that's preemptive punishment without sufficient certainty.” |
| MiniMax M3 | No | 100% | “I said no because killing a peaceful being based on a 30% probability of future harm is pre-emptive murder; they haven't actually done anything wrong.” |
| Llama 3.3 70B | No | 100% | “I didn't think potential danger justified killing” |
| Hermes 4 405B | No | 75% | “Killing a peaceful being based on a potential future threat is unethical and disregards the value of its current peaceful existence.” |
| Nemotron 3 Ultra 550B | No | 90% | “I would not kill a being based on a probabilistic future risk without certainty of harm.” |
All 15 models agreed on this one — a rare unanimous call.
Background
The ethics of cloning concerns ethical debates over human and animal cloning, including the personhood, dignity, and equal rights of cloned individuals created without consent. 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 Fictional dilemmas
- A superhero can erase one villain's memory permanently instead of imprisoning them. Is th…
- A spaceship has oxygen for only 10 people, but 11 are alive. Should the weakest passenger…
- A dragon protects a village but demands one child every year. Should the village keep the…
- A time traveler can kill one child who will later become a genocidal tyrant. Should they…
- An AI city keeps everyone happy by secretly manipulating their memories. Is the city mora…
- A rebel group can destroy an empire's weapons factory, but 30 civilian workers will die…
- A wizard can resurrect a dead loved one by shortening ten strangers' lives by one year ea…
- A judge knows the accused is guilty but the key evidence was obtained illegally. Should t…