Library

GLOSSOPETRAE

GLOSSOPETRAE: The engine that builds a language no AI has ever seen

An engine that builds complete new languages from a single number, then measures how fast AI models learn them and what can be hidden inside.

What this means for you

Every language you have ever heard took centuries to form. GLOSSOPETRAE makes one in under a second. You give it a number. It gives you back a complete language that has never existed: its own sounds, its own grammar, its own dictionary, its own alphabet drawn as shapes you can look at, and a voice that can pronounce it. Give it the same number tomorrow and you get the same language. Change one digit and you get a different one.

GLOSSOPETRAE after forging a language, showing its generated grammar summary and quick reference
Fig. IA language forged from one seed number a moment earlier. The panel on the right is the crib sheet an AI model is given to learn it.

That would be a curiosity if it stopped there. The reason it matters is what the project did next. It used these brand-new languages to run experiments on AI models that would be impossible with real ones.

Imagine wanting to know how quickly a child picks up a language, but every child you test has already overheard some French. You can never be sure what they learned from you. So you invent a language the night before, one that nobody on earth has heard, and teach that. Now whatever they know, they learned in the room.

AI models have read most of the internet, so there is almost no real language they have not seen. An invented one is a clean slate. That let the researchers ask three questions.

What they found

How fast can a model learn a language it has never seen? Very fast, and in an odd way. Shown no examples, one example, or three, the models tested got nothing right. Shown eight, most of them translated almost perfectly. There was no gentle slope between the two. The project calls it a cliff, and reports the same shape across nine different leading models.

The CliffΛΥΣΙΟΣ · LIBERHow fast a model learns a language it has never seensemenXLIIC0nihil0nihilInihilIIIVIIIIA NUMBERThe seedIIA NEW LANGUAGENever seen by anyoneIIITHE CLIFFExamples shown: none, one, three, eight
Fig. IIA language made from one number, then taught by example. With none, one or three examples the models tested got nothing right. With eight, nearly everything.

Does a language need to be readable by people to be usable by a machine? No, and the result ran the opposite way from what you would expect. When the invented language was made harder for humans to read, the models got better at using it. A language can be built for machines that people cannot follow at all.

Can two AI systems pass a note that a third AI, set to watch them, will miss? Yes. Different AI models break text into pieces differently before they read it. Some quietly drop certain invisible characters. Others keep them. A message written in characters the watcher drops and the receiver keeps goes straight through. The watcher is not careless. It never receives the message to begin with.

The Blind WatcherΛΥΣΙΟΣ · LIBEROne message, read by two machinesEpistulaconstituitibinarrarepraefatioverissimadumnugasessealiquidnovitatemdarenovisconstituitibinarrarepraefatioverissimadumnugasessealiquidnovitatemdarenovisnihilconstituitibinarrarepraefatioverissimadumnugasessealiquidnovitatemdarenovisavecustossociusITHE MESSAGEPlain words, hidden marks betweenIITHE WATCHERIts reader drops the marksIIITHE RECEIVERIts reader keeps them
Fig. IIIWhy the watcher misses it. Its reader throws the hidden marks away before the model ever looks, so the message it sees is clean. The receiver’s reader keeps them.

That last finding has a practical edge. A common safety idea is to have one AI supervise another, often one from a different company for independence. The project reports that this mixed setup is where the blind spots were richest. It also found a harder version of the problem: messages hidden in ordinary word choices, with no invisible characters at all, slipped past every watcher tested most of the time. The authors say plainly that there is no quick fix for that one.

A bar chart from GLOSSOPETRAE of how many character categories each of twelve AI models cannot see
Fig. IVThe project’s own blind-spot chart: how many categories of hidden character each of twelve models fails to see. Models from the same family share the same gaps.

How to use it

The whole project is free on GitHub at the link above, and you can read it without an account. To play with the language engine, go to elder-plinius.github.io/GLOSSOPETRAE in a web browser. There is nothing to install. You can also download the project and open the file called index.html for your own copy. Type any number as a seed and explore the language it makes: read its grammar, see its script, hear a word spoken, or age it by a few centuries and watch the sounds drift.

The research is written up as a paper inside the project, with the raw results of every experiment included so that anyone can check the figures. The authors note that their own first draft contained errors, which they caught by auditing themselves, and they left the account of that in the paper. If you only read one thing, read that section. It is a good example of how this kind of work should be shown.