Personal Visual Reasoning Test Shows AI Far From AGI

adam_dorr · x · 2026-07-10

Developer Adam Dorr possesses a personal benchmark test for frontier AI models that no model has passed yet, including GPT 5.5, Opus 4.8, and Grok 4.5. The test features no related training data, making it an out-of-distribution problem that an average 12-year-old can easily solve. This indicates that current AI systems are still quite far from achieving true AGI.

Original post →

More from AGI Musings

AGI Musings channel →