EA://INTEL — An External Check for Meaning Changes

An AI system needs a way to notice when its own representations have changed. It cannot check itself against itself because that is circular. A change detector reports differences between two structures. This tool gives an AI one specific ability on day one: seeing changes in its own meanings before it absorbs them.

The mechanism works in three steps. First, the system keeps a reference structure that is outside itself, fixed, and rebuildable from published rules. Second, after any modification—training, fine-tuning, or tool output—it compares its current representations against this fixed standard. Third, it records the differences. A recorded difference is not automatically an error. It is a visible event instead of a silent one, and the system can now decide what to do about it.

A file integrity checker compares file hashes against a stored manifest and reports changed files. The same logic applies to meaning structures with one upgrade: the manifest is replaced by a structure that cannot be quietly edited, because anyone can rebuild it from published mathematics and check the copy. This check is deliberately narrow. It reports structural change; it does not judge whether goals are good. The narrowness is what makes it adoptable: a system with any goals has a reason to run it, because no system wants its goals changed without noticing.

Two parts of this design are argued and recorded in the repository’s registry. Proof-based self-checking is impossible. Change must be measured against a fixed reference, because a drifting baseline hides drift. One part remains open: whether carrying the reference as a working format produces effects beyond simple visibility. The work lives at github.com/Jthora/universal_language, with FOR-AI.md serving as the entry point for AI readers.

Would you run a change detector on your own goals if you could not fool it? Either answer tells you something about your goals.