Two Models Read Doodles: One Guesses, One Refuses
goodside · x · 2026-07-12
The author tested models' handwriting recognition capabilities using a meaningless doodle drawn "randomly on a phone with eyes closed":
- GPT-5.6 Sol consistently fabricates an answer, exhibiting hallucinated recognition.
- Claude Fable 5 admits it cannot read the text, even arguing that it doesn't look like writing at all.
The author notes that readers can test it themselves using the image provided in the post.
Related event: GPT-5.6 Sol vs Claude Fable 5 Doodle Test Sparks Hallucination Debate(5 posts)→
More from Fun
- Why do bullshitters climb the corporate ladder? Hot air rises — nabla_theta · 2026-10-27
- Open-source decision models sold their own base in Red Alert 2 — then learned until no human could beat them — heathercmiller · 2026-10-27
- Dev jokes: predict Fable 5.5's release every day and you'll eventually be right — majidmanzarpour · 2026-10-27
- "We built the altar first": striking AI-generated monumental sci-fi imagery — piotrbinkowski · 2026-10-27
- AI Infra Meme: While Countries Need TW-Scale Power for Datacenters, People Argue About Lightbulbs — mgostIH · 2026-10-27
- Von Neumann's 1947 math quote resurfaces alongside rare 1952 IAS computer photo — pickover · 2026-10-27