Humans Outperform AI in Learning Through Experience on Complex Benchmarks
joshua_saxe · x · 2026-08-29
Research by EpochAI indicates that while AI can one-shot many tasks, repeatedly trying a task it fails at doesn't help—lacking effective learning trajectories. In contrast, humans playing the complex board game Earthborne Rangers mastered it after just 5 playthroughs, reaching a level no evaluated AI has ever achieved, highlighting humans' unique ability to improve with experience.
Related event: Epoch AI study finds AI lacks human-like learning through repeated attempts(2 posts)→
More from Research
- Anthropic shows AI researchers autonomously improving alignment of other models — VraserX · 2026-08-30
- Learn Positional Encodings derivation from first principles — zainhas · 2026-08-30
- COLM Paper Traces Capability Provenance in LLMs via Gradient Attribution — ziv_ravid · 2026-08-30
- Toby Ord paper argues recursive self-improvement has physical limits — Exponential View (Azeem Azhar) · 2026-08-30
- AI Formalization Tools Fable and Sol Spot First Repairable Error in Published Literature — Sauers_ · 2026-08-30
- Mark Schmidt Posts ICML Tutorial Video: Is Numerical Optimization Theory Irrelevant to ML Practice in 2026? — MarkSchmidtUBC · 2026-08-30