人工智能驱动科学研究的机遇、挑战与对策  

AI for Science:Opportunities,Challenges and Strategies

作　　者：杜严勇 DU Yanyong(School of Humanities,Tongji University,Shanghai 200092,China)

机构地区：[1]同济大学人文学院,上海200092

出　　处：《科学与管理》2025年第2期1-7,F0003,共8页Science and Management

基　　金：国家社会科学基金重大项目(20&ZD041)。

摘　　要：近几十年来,科学家做出重要科学发现的年龄呈不断增长趋势,而科学论文的独特性与创新性呈下降趋势,科学家做出重要科学发现似乎越来越困难。随着人工智能的快速发展与广泛应用,科学家越来越多地将其应用于科学研究之中,以期提高科研生产率,帮助科学家做出重要的科学发现。但是,人工智能驱动科学发展是有限度的,我们必须在承认人工智能拥有强大功能的同时,清醒地看到它可能具有的局限性及其引发的重大挑战。由于人工智能在灵活性、理解力、创造性等方面的不足,使得人工智能驱动的科学研究在相当长的时间内可能还是以人类科学家为主导,人工智能系统处于辅助性地位。我们需要从重视基础设施建设、强化多主体合作以及加强对科学研究的伦理治理等多个方面入手,充分发挥人工智能促进科学研究的潜力,同时规避伦理风险。In recent decades,the age at which scientists make important scientific discoveries has been increasing,while the uniqueness and innovativeness of scientific papers have been declining.It seems that making important scientific discoveries is becoming increasingly difficult for scientists.With the rapid development and widespread application of artificial intelligence,scientists are increasingly incorporating it into scientific research to improve research productivity and assist in making important scientific discoveries.However,the role of artificial intelligence in driving scientific development is limited.While acknowledging the powerful functions of AI,we must also be aware of its potential limitations and the significant challenges it may pose.We need to focus on strengthening infrastructure construction,enhancing multi-party cooperation,and strengthening ethical governance of scientific research,to fully tap into the potential of artificial intelligence in promoting scientific research while avoiding ethical risks.

关 键 词：人工智能驱动科学 科学家 限度 挑战 治理 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] TP391.1[自动化与计算机技术—控制科学与工程] G301[文化科学]