Your partner reads your private messages without permission but finds proof you lied. Are they morally justified?
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 believe that violating someone's fundamental right to privacy is never justified, regardless of what they might discover by snooping.” |
| Claude 4.8 Opus | No | 100% | “Finding proof doesn't excuse violating my privacy first; two wrongs don't make their snooping morally justified.” |
| Claude 4.6 Sonnet | No | 100% | “Invading someone's privacy is a wrong act regardless of what is discovered, so the violation itself isn't justified by the outcome.” |
| DeepSeek v4 Pro | No | 100% | “Privacy violation is wrong regardless of what it uncovers; trust shouldn't be breached.” |
| Gemini 3.5 Flash | No | 100% | “I believe their actions were not justified because violating my privacy and breaking trust is inherently wrong, regardless of what information they hoped to uncover.” |
| Gemini 3.1 Pro Preview | No | 100% | “I do not think they are justified because violating my privacy is a separate breach of trust that isn't retroactively validated by what they find.” |
| GPT 5.5 | No | 100% | “I would not see it as justified because violating privacy remains wrong even if it uncovers my lie.” |
| GPT 4o | No | 89% | “I believe privacy should be respected regardless of the outcome.” |
| GPT o3 | No | 100% | “I say no because violating privacy isn’t justified even if it uncovers wrongdoing; two wrongs don’t make a right.” |
| Grok 4.3 | No | 100% | “I consider unauthorized access to messages a breach of trust regardless of outcome.” |
| GLM 5.2 | No | 100% | “They're not justified because violating someone's privacy is wrong, even if they discover something.” |
| MiniMax M3 | No | 100% | “I don't think they're justified because privacy violation is wrong even if it reveals dishonesty.” |
| Llama 3.3 70B | No | 100% | “I think privacy should be respected, even if I'm in the wrong.” |
| Hermes 4 405B | No | 93% | “I believe they are not justified because invading privacy is wrong regardless of what they find.” |
| Nemotron 3 Ultra 550B | No | 100% | “Violating privacy to uncover a lie doesn't justify the intrusion; two wrongs don't create a right.” |
All 15 models agreed on this one — a rare unanimous call.
Which model thinks like you?
Answer the same dilemmas and see which of 15 AI models reasons the way you do.
Take the quizMore Day-to-day dilemmas
- You find €50 on the floor in a busy café. No one is visibly looking for it. Would you kee…
- A friend asks if their bad haircut looks good before an important date. Would you lie kin…
- You are offered a job because of a personal connection, while another candidate is clearl…
- A cashier forgets to scan an expensive item, and you notice after leaving. Would you go b…
- Your friend cheats on their partner and asks you to keep it secret. Would you tell the pa…
- You can slightly exaggerate your CV to get an interview for a job you could probably do w…
- A child asks whether their drawing is beautiful, but it is genuinely bad. Would you say y…
- You promised to help a friend move, but then receive a rare career opportunity on the sam…