Conceptual Limitations
摘要
If the TuringTuring testTuring test would be used as benchmark for the successful implementation of artificial intelligenceArtificial Intelligence (AI), statisticsStatistics-based AI is in a dilemma—at least as long as it is still operating in the in black boxBlack box mode. This is because one only need to follow up on a question, which can be answered as well as possible by the AI, with the next question: “Why?”.