中国科学院院士鄂维南 未来民科是科研的主力 AI 时代每个有想法有能力的团队和个人都可以参与到科技创新中去CAS Academician E Weinan: Citizen Scientists Will Be a Main Force in Future Research


AI free token/AI免费积分
中国科学院院士鄂维南 未来民科是科研的主力 AI 时代每个有想法有能力的团队和个人都可以参与到科技创新中去

10 月 17 日,在光华思想力论坛上,中国科学院院士、北京大学讲席教授鄂维南表示,人工智能本质上是“数学 + 计算机工程”,将大幅降低科研门槛,推动“科研平权化”。“将来民科是科研的主力”,他预测,借助开放实验室和云服务,“每一个有想法有能力的团队和个人都可以参与到科技创新当中去”。

鄂维南指出,企业真正做生产时缺乏原始创新,只能在管理、售价等维度竞争,“内卷的根本原因是创新能力不够”。他认为,高端工业软件、高端机床数控平台当下受制于人,但“现在不不过可以比较短的时间内达成自主可控,甚至可以‘换道超车’”,可构建 AI 原生工业软件体系,“原则上几乎可以拿出不但可以对标西门子,甚至可以超越西门子的体系,一两年内就可以做成”。

他区分 AI for Science 与 AI for Industry,指出后者起步更难,因工业软件更封闭、工业数据大部分闭源,但工业生产线建模因为宏观较容易,将来核心任务是把工业生产线建模做到位。他判断这一进程“三年、最多五年就可以达成”。

鄂维南同时警示,人工智能时代现行的科研体系将会面临巨大的挑战。他以数学界为例称,若四大顶刊投稿量增加百倍,“笔者实际上并不知道如何应对”。他强调颠覆性创新“很大程度上靠自由探索而达成的,而不是任务定义而驱动的”,并批评“把自由探索和写论文等同于一起,这是所犯的错误”。

他认为,科研成果形态将从论文转向知识产权,产业形态也将重构,高校必须构建深入理解需要的创新驱动一体化人才培养体系。

本文来源:https://www.ithome.com/1/014/371.htm
若有版权问题,请来信告知:13951431913#139.com 或在右方提交评论。

On October 17, at the Guanghua Thought Forum, E Weinan, an academician of the Chinese Academy of Sciences and a professor at Peking University, said that artificial intelligence is essentially "mathematics + computer engineering", which will significantly reduce the threshold of scientific research and promote "equal rights in scientific research". "People's science is the main force of scientific research in the future", he predicts that with the help of open laboratories and cloud services, "every team and individual with ideas and abilities can participate in scientific and technological innovation".

E Weinan pointed out that enterprises really lack original innovation when doing production, and can only compete in management, selling price and other dimensions. "The root cause of inner roll is insufficient innovation ability". He believes that high-end industrial software and high-end machine tool CNC platforms are currently subject to people, but "now it is possible to achieve autonomous control in a relatively short period of time, and even ‘change lanes and overtake‘", which can build an AI native industrial software system. "In principle, it is almost possible to come up with a system that can not only benchmark Siemens, but even surpass Siemens, and can be done within one or two years."

He distinguished between AI for Science and AI for Industry, and pointed out that the latter is more difficult to start with, because industrial software is more closed and industrial data is mostly closed-source, but industrial production line modeling is easier because it is macroscopic, and the core task in the future is to model industrial production lines in place. He judged the process to be “achievable in three years, at most five years”.

At the same time, E Weinan warned that the current scientific research system in the era of artificial intelligence will face great challenges. Taking the mathematical community as an example, if the number of submissions to the top four journals increases by a hundred times, "I actually don't know how to deal with it". He emphasized that disruptive innovations “are largely driven by free exploration rather than task definition,” and criticized “the mistake of equating free exploration with writing a thesis.”

CAS Academician E Weinan: Citizen Scientists Will Be a Main Force in Future Research

He believes that the form of scientific research achievements will shift from papers to intellectual property rights, and the industrial form will also be reconstructed. Universities must build an innovation-driven integrated talent training system with a deep understanding of the needs.

This article is from: https://www.ithome.com/1/014/371.htm For copyright issues, please write to: 13951431913#139.com or submit a review on the right.

暂无评论,快来发表第一条评论吧!No comments yet. Be the first!

📮 需求咨询📮 Inquiry