Reddit debates which benchmarks humans still win as AI tests saturate

Reddit users discuss which major benchmarks humans still outperform frontier LLMs on, as tests like SimpleBench and ARC-AGI saturate, sparking debate over how AGI should be defined.

2026-11-11 ~ 2026-11-11 · 2 related posts

• Which benchmarks are left? Redditors weigh in as ARC-AGI and knowledge tests saturate — ImproveOurWorld · 2026-11-11
• Which benchmarks do humans still win against frontier AI? Reddit debates AGI criteria — ImproveOurWorld · 2026-11-11