Intelligence doesn't have a lot of scientific ground either. It's pretty hard to define what intelligence is, or at least have a scientific definition that is p

Intelligence doesn't have a lot of scientific ground either. It's pretty hard to define what intelligence is, or at least have a scientific definition that is precise enough. The Turing Test is only a measure, it doesn't help to reach a definition. Practical research will always hit a ceiling if scientists cannot try to define what they're looking for. Even machine learning is not a good definition. There are other attemps, like "sophisticated statistics" or "statistical prediction". Kudos for this tweet.