how clanker
are you?
// a surprisal Turing test
Drop an @handle — we pull their recent posts and grade how clanker (AI-like) they write. Language models predict the next token; the more predictable the writing, the more clanker.
→ or diagnose yourself — finish 5 sentences
⚠ demo mode: inference isn't funded yet — scores come from a deterministic stand-in, not the real models.
pulling recent posts…
- ⋯ finding the account on X
- ⋯ pulling their 5 most recent posts (retweets & replies excluded)
- ⋯ scoring word-by-word vs Llama · DeepSeek · Qwen3
~a minute or two — we walk their posts token by token, and the models don't rush
▌
3–10 words. be yourself. or don't.
interrogating the models…
token by token. they can't hide their logprobs.
you are clanker
one square per word · green = human · red = clanker
deeper stats
for fun, from public posts — not a judgment of the person. remove me
method: each word you typed is scored by its surprisal under each model — −log pmodel(word), from the model's top-20 next-token probabilities conditioned on your text so far. mean surprisal in nats over your words is how predictable you were; low = clanker. words outside the top-20 are floored, scores are normalized against each model's self-baseline, and your overall is your nearest (least-surprised) model. (equivalently: the KL from your one-hot word choice to the model's distribution collapses to exactly this surprisal.) (demo mode: logprobs are currently simulated.)