Ai Second Opinion

AI Second Opinion

[](https://github.com/igorsaevets/ai-second-opinion/actions/workflows/selftest.yml) [](LICENSE) [](INSTALL.md)

**One AI agreeing with you proves nothing. Three of them arguing is worth reading.**

Send the same document to three different top-tier AI models at once. Get back what each one found, where they contradict each other — and a mechanical check of whether they actually did the research or quietly made it up.

[Русская версия](README.ru.md) · [How it works, in technical detail](TECHNICAL.md) · [Install](INSTALL.md) · [When something breaks](TROUBLESHOOTING.md) · [For AI agents](AGENTS.md)

---

The problem this solves

You ask an AI to review your strategy memo. It tells you the memo is strong, adds three supportive points, and cites four sources.

That answer is nearly worthless, for three reasons most people never check:

1. It is built to agree with you. You wrote the memo, you asked the question, and the model optimises for a helpful-feeling reply. Ask the same model to attack the memo and it will find problems it just told you did not exist.
2. One model has one set of blind spots. Whatever it was weak at yesterday, it is weak at today, and nothing in its answer tells you which parts those are.
3. The sources may not exist. Models generate citations that look right — real domain, plausible path, correct-sounding document number — for pages that were never opened and sometimes never existed. In one measured run here, a model produced 11 source links; 3 of them were dead, and its conclusions were still correct. That combination is the dangerous one, because it survives a casual read.

What this does instead

• **Three independent models, same do