alt Hacker NewsOr they just point to the turing test which was the defacto standard test for something so nebulous. And behold: LLM can pass the turing test. So they think. Can you come up with something better (than the turing test)?
Or they just point to the turing test which was the defacto standard test for something so nebulous. And behold: LLM can pass the turing test. So they think. Can you come up with something better (than the turing test)?
But the Turing test (which I concede, LLMs do pass) doesn't test if some system is thinking; it tests if the system can convince an unbiased observer that it is thinking. I cannot come up with a better "is this thing thinking" test, but that doesn't mean that such a test can't exist; I'm sure there are much smarter people then me trying to solve this problem.