Until <5 years ago, AI was almost entirely a purely academic field, theoretical at that. Those same academics admit themselves that they're surprised at how we

Until <5 years ago, AI was almost entirely a purely academic field, theoretical at that. Those same academics admit themselves that they're surprised at how well LLMs do considering how simple(?) rudimentary(?) the logic underneath is. I don't quite understand what you're saying. That these academics were being lazy by not properly investigating/publishing their findings? That doesn't seem right.