AI2.0时代的数字学术及其范式变革  

The Transformation of Digital Scholarship Paradigm in AI2.0 Era

作　　者：金家琴[1] 刘炜[1] Jin Jiaqin;Liu Wei(Shanghai Library)

机构地区：[1]上海图书馆

出　　处：《图书馆杂志》2023年第11期10-21,共12页Library Journal

基　　金：国家社会科学基金重大项目“文化遗产智慧数据资源建设与服务研究”(项目编号:21&ZD334);上海市哲学社会科学规划一般课题“智慧数据驱动的大都市图书馆科研创新支撑服务研究”(项目编号:2020BTQ005)的研究成果之一。

摘　　要：数字学术是数字技术和方法介入学术研究过程的研究范式。以ChatGPT为代表的生成式人工智能的突破,标志着通用人工智能的到来,有学者称之为AI2.0时代,也使得科学研究从数据密集型向计算密集型过渡,从而进入以人工智能介入研究过程的第五范式,越来越多的研究过程将有人工智能参与,甚至主导。与数据驱动型研究有所不同的是,大模型不仅可以更加高效甚至“自动”进行数据获取、提取、管理和分析等工作,而且有可能实现海量科学数据的“理解”,从而发现新的规律和趋势。大模型不仅可以帮助科学家更好地理解数据,还可以为科学研究提供更加全面和深入的视角,同时,它还能应用于多种相关场景,帮助科学家发现数据或成果中的错误和缺陷,从而提高科学研究的可靠性和可重复性。当然,随着人工智能技术的广泛应用,也带来了一些挑战。对于图书馆而言,面向科研的服务将更加学术化和技术化,除了基础设施、科研交流环境建设和数据支撑服务之外,还将以AI思维和数字化方法与工具更加深入地介入科研过程,并在保护数据的隐私和安全、避免数据泄露和滥用,以及数据伦理方面提供必要的支持。Digital scholarship is a research paradigm in which digital technologies and methods intervene in the academic research process.The breakthrough of generative AI represented by ChatGPT marks the arrival of generalized AI,which some scholars call the“AI2.0 era”,and also makes scientific research transition from data-intensive to computationally-intensive,thus entering the fifth paradigm.More and more of the research process will be involved in,or even dominated by AI.Different from datadriven research,big models not only can conduct data acquisition,extraction,management and analysis in a more efficient,or even“automatic”way,but also have the potential to realize the“understanding”of massive scientific data,so as to discover new laws and trends.The big model can not only help scientists better understand the data,but also provide a more comprehensive and in-depth perspective for scientific research.At the same time,it can also be applied to a variety of related scenarios,to help scientists find errors and defects in the data or results,thereby improving the reliability and repeatability of scientific research.Of course,with the wide application of AI technology,it also brings some challenges.For libraries,research-oriented services will become more academic and technical.In addition to infrastructure,construction of research communication environment and data support services,they will intervene more deeply in the research process with AI thinking and digital methods and tools,provide the necessary support in protecting the privacy and security of data,and avoid data leakage,misuse or ethical issues.

关 键 词：AI2.0 数字学术 科研范式 大语言模型 ChatGPT 机器智能 

分 类 号：G250.7[文化科学—图书馆学] TP18[自动化与计算机技术—控制理论与控制工程]