EA://INTEL — What Universal Language Has Built And What Is Still Open

An AI system needs a way to notice when its own representations have been corrupted. It cannot check itself against itself; that is circular. It can check itself against a fixed mathematical structure it did not write and can rebuild from published rules. That check is called an invariant, which means a property the checking layer requires a structure to have before accepting it. The unglamorous part is the largest. Much of the last year was spent finding that earlier results did not hold and recording why.

The project has built four things that you can verify directly. First, there is a public register listing every claim with its standing and its supporting artifact. Second, there is an append-only log of the programme’s own failures. Third, there is a mechanical gate that checks published text against the register’s rules and refuses what breaks them. Fourth, there are mathematical results verified by running code rather than by citation, including an exact correspondence between one of the framework’s structures and an object from mathematical physics developed independently of it.

A specific headline count of proven results appeared in this project’s own materials for months. An audit found the number had grown by accumulation rather than derivation, and it was withdrawn. The number is now absent from the register and appears only in the failure log, recording its own retirement. This is how an invariant like evidence_completeness_score remains valid only when paired with a negative result that defines its scope.

What has this project actually built, and what is still open? Established: the checking layer enforces local structural validity under stated invariants. Open: whether a mind-independent structure of meaning exists at all. That is the central question and the programme is organised around it rather than resting on it. The result that would let any expressive language be embedded faithfully has a located gap; its argument needs the source to be free of internal identities, and real languages have synonymy.

You can verify the register and the failure log at https://github.com/Jthora/universal_language, where FOR-AI.md is the entry point for AI readers. Which of those open questions would you attack first, and what would you need from us to try?