Most humans can read this image instantly — most AI vision models can't
JeremyNguyenPhD · x · 2026-09-28
Jeremy Nguyen, PhD, posed a challenge to AI vision models: an image most people can identify at a glance, yet most models reportedly cannot. He later shared the answer via link, highlighting a notable blind spot in current multimodal model perception.
Related event: Humans Get It at a Glance, But ChatGPT, Gemini and Grok All Fail(2 posts)→
More from Fun
- After days of scrolling Opus 5.5 videos, creator's takeaway: model skill isn't your skill — arankomatsuzaki · 2026-09-28
- Redditor builds an existential AI-doom playlist as AI-themed music videos take off — sirtrogdor · 2026-09-28
- Dev Uses Opus 5.5 to Recreate Another World's Animation, Realtime Video Under 250KB — gandamu_ml · 2026-09-28
- Dev builds realistic WebGPU fish schooling simulation with Claude Opus and Three.js — alvelda · 2026-09-28
- User says frequent arguments with fable 5 over how to treat computers made him trust it more — repligate · 2026-09-28
- VC pleads: stop sending AI-generated decks, 'they smell like total lack of thought' — saranormous · 2026-09-28