Quality medical care reaches the rich, and the rich countries. This is an attempt to change that — to put a structured, honest, multilingual clinical reasoning tool in the hands of everyone, for nearly nothing.
Much of the world lives under fewer than 1 physician per 10,000 people.
For billions, "see a doctor" is not an option that exists. And even where doctors exist, medicine is often dephased — out of step with the person in front of it: rushed, language-barriered, priced out, or shaped by what is billable rather than what is wrong. The knowledge to reason through a health problem is not scarce. Access to it is — deliberately, because gatekept knowledge is profitable and free knowledge is not.
Free, structured health education — the same clinical intake (OPQRST) physicians are trained on, run by multiple AIs that debate to consensus and must cite guidelines, in the person's own language, ending in a one-page summary they can carry to any doctor. It is not a doctor and never pretends to be one. Where a physician exists, the human does 100% of the verification. Where none exists, a structured second opinion is still vastly better than guessing alone.
Built and released by one board-certified family doctor who automated his own practice and paid for all of it out of pocket. Either people get involved — use it, fork it, build on it — or it dies on its own. That is why it is open-source (MIT) and free forever.
So none of this can be quietly rewritten or buried, everything published is recorded on a hash-chained, append-only public ledger — each entry signed and linked to the one before it, so a single altered or deleted record breaks the chain visibly. It is the opposite of a system that profits when you can't check its work.
→ read the ledger: instanthpi.ai/4chan · run your own node: the modules