Which benchmarks are left? Redditors weigh in as ARC-AGI and knowledge tests saturate
ImproveOurWorld · reddit · 2026-11-11
A Reddit thread asks which benchmarks remain unsaturated now that SimpleBench, ARC-AGI, and most knowledge benchmarks have been solved. The author names Remote Labor Index, HLE-Diamond, Millennium Prize problem solutions, and robotics benchmarks as the main ones left.
The thread also poses open questions: when will humans run out of new benchmarks, and is the 'singularity' precisely the moment benchmark design itself gets surpassed?
Related event: Reddit debates which benchmarks humans still win as AI tests saturate(2 posts)→
More from AGI Musings
- Dictating notes to an AI agent feels as mundane as putting on shoes — _FelixSimon_ · 2026-11-11
- Data center blocked over water-poisoning myth cost community billions in taxes — Dan_Jeffries1 · 2026-11-11
- AI agents are becoming software buyers, and startups aren't ready — BakingBreadBB2 · 2026-11-11
- Reddit essay: what we fear isn't AI but Artificial Life — agents that want to survive — Honest_Science · 2026-11-11
- Against p(doom): assigning probabilities to one-shot events is sillier than it sounds — GabGarrett · 2026-11-11
- Altman says cancer and Alzheimer's will be 'relics' within a decade; meme tallies his $50k outfit — morqon · 2026-11-11