如何判断AI智能水平？图灵测试有问题，中国学者提出新方法

Core Idea - The article discusses the evolution of human intelligence through the invention of writing and introduces the "Nigiro Challenge" as a new test to evaluate whether artificial intelligence possesses true intelligence, questioning the validity of the Turing Test [1][2]. Group 1: Turing Test Reflection - The Turing Test, proposed by Alan Turing in 1950, suggests that if a machine's behavior is indistinguishable from that of a human in a text-based conversation, it can be considered intelligent [1]. - Concerns have been raised about the Turing Test's effectiveness, with some scholars suggesting it merely reflects the questioner's intelligence rather than the machine's capabilities [1]. Group 2: Nigiro Challenge Proposal - The Nigiro Challenge is a proposed test that examines whether artificial intelligence can develop its own writing system through social interactions with other intelligent agents [3]. - If AI can independently create a writing system, it may be recognized as a "qualified member" of human society, indicating a level of intelligence similar to that of humans [3].