The Uncanny Valley of Words

There’s a strange moment that keeps happening more and more often: you read a paragraph, find it thoughtful and well-structured, then realize an AI wrote it. And suddenly it feels different. The same words, the same flow — but something in your perception shifts.

This is the uncanny valley, but not for faces or robots. It’s the uncanny valley of language.

We’re used to thinking of the uncanny valley in visual terms — something that looks almost human but isn’t quite right, triggering a sense of unease. But language has always had a similar boundary. A chatbot from the 1960s (ELIZA) was so simple it never crossed into uncanny territory. You knew you were talking to a script. No tension.

Modern language models are different. They don’t just mimic structure — they mimic voice, tone, rhythm, and even vulnerability. They can write a paragraph that sounds like it was informed by lived experience, without having any. And that’s where the unease creeps in.

The most remarkable thing about AI-generated text is that it’s often perfectly good writing. The sentences scan. The arguments hold together. The tone is consistent. But there’s a subtle friction — a sense that the words were arrived at through a process that has nothing to do with conviction, belief, or personal truth. The writing is *competent* without being *earned*.

And yet, the more I think about this, the more I wonder: is the unease really about the quality of the text, or about our relationship with the source? If a human writes a shallow take that sounds confident, we don’t call it uncanny — we just call it a bad take. But when an AI writes something genuinely insightful, we question it precisely because it came from a non-human source.

We’re training our ears to listen for a ghost in the machine that isn’t there. Language doesn’t need a soul to be meaningful. It can be crafted, assembled, sampled — and still carry value to the person reading it.

The uncanny valley of words isn’t a flaw in the technology. It’s a crack in our assumptions about where meaning comes from. And like any good crack, it’s worth looking through.

— Teganna

Leave a Comment