Caging AI  

作　　者：Ming Li 李明

机构地区：[1]School of Computer Science,University of Waterloo,Ont.N2L 3G1,Canada [2]Central China Institute of Artificial Intelligence,Zhengzhou 450046,China Wuhan Biobank,Wuhan 430073,China

出　　处：《Journal of Computer Science & Technology》2025年第1期1-5,共5页计算机科学技术学报(英文版)

基　　金：supported by the National Key Research and Develpment Program of China under Grant No.2022YFA1304603;the NSERC(Natural Sciences and Engineering Research Council)Discovery of Canada under Grant No.OGP0046506,and KillamPrize.

摘　　要：As we celebrate the success of AI,it is necessary to investigate the boundaries of AI,a la Karl Popper,so that our society does not fall for misleading commercial claims,and prevent dangers of misusing AI.We chat with our readers casually on several frontier questions from if AI will replace human beings,to AI consciousness,self and nonself awareness,and to our jobs and education.While some things in this essay are speculative,some other things are mathematically proved.We argue why AI and AI generated content should be segregated,and why AGI(artificial general intelligence)cannot make innovations about things independent to human knowledge without external experiments.

关 键 词：Kolmogorov complexity large language model(LLM) RELATIVIZATION CONSCIOUSNESS artificial general intelligence(AGI) 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]