I appreciate this reflection on why ARC matters, but in the replies I expressed some reservations about what high accuracy on ARC benchmarks actually reflects for assessing "human-like fluid intelligence" (Chollet's stated goal for ARC).
2020s: AI "System Prompts": lengthy, carefully constructed sets of expert rules for a particular domain, created by "prompt engineers" 1980s: AI "Expert Systems": lengthy, carefully constructed sets of expert rules for a particular domain, created by "knowledge...