The Logic,Risks and Governance of Artificial Intelligence-Driven Scientific Research
Xu Dongbo1,2
1. School of Marxism,Southeast University,Nanjing 211189,China; 2. Resarch Institute for the Development of Socialism with Chinese Characteristics,Southeast University,Nanjing 211189,China
Abstract：Artificial intelligence based on disruptive innovation technology has a unique research-driven logic:to improve the efficiency of scientific research through catalytic drive,to promote multi-dimensional integration and interaction through collaborative drive,and to break research path dependence through algorithmic drive.The essence of triple drive logic is a formal and programmatic empowerment.Whether science can achieve substantial innovation depends on the innovation potential of researchers,artificial intelligence,and the level of coupling and interaction between the two.While artificial intelligence endows scientific research with new prospects,it also implies multiple potential risks,such as“moral misconduct”in scientific research actions,“mechanical concealment”in research processes,“fuzzy authenticity”in formal innovation,and“no one bears academic responsibility”.In order to reduce driving risks and improve driving resilience,it is necessary to regulate the development and application of artificial intelligence systems,guide researchers to carry out responsible innovation,and promote deep collaborative interaction between scientists and artificial intelligence.
[1]王飞跃,缪青海.人工智能驱动的科学研究新范式:从AI4S到智能科学[J].中国科学院院刊,2023,38(4):536-540. [2]VAN DIS E A M,BOLLEN J,ZUIDEMA W,et al.ChatGPT:five priorities for research[J].Nature,2023,614(7947):224-226. [3]BOSTROM N.Get ready for the dawn of superintelligence[J].New Scientist,2014,223(2976):26-27. [4]中华人民共和国中央人民政府.科技部启动“人工智能驱动的科学研究”专项部署工作[EB/OL].(2023-03-27)[2023-06-06].https://www.gov.cn/xinwen/2023-03/27/content_5748495.htm. [5]UNESCO.Recommendation on the ethics of artificial intelligence[EB/OL].(2021-11-24)[2023-06-06].https://unesdoc.unesco.org/ark:/48223/pf0000379920.page=14. [6]郦全民.人工智能在生产力中的角色[J].华东师范大学学报(哲学社会科学版),2023,55(5):6-12,170. [7]BOSTROM N,YUDKOWSKY E.The ethics of artificial intelligence[M]//FRANKISH K,RAMSEY W M.The Cambridge handbook of artificial intelligence.Cambridge:Cambridge University Press,2014. [8]LELE A,LELE B.Disruptive technologies for the militaries and security[M].Singapore:Springer,2019. [9]LINKOV I,TRUMP B D,POINSATTE-JONES K,et al.Governance strategies for a sustainable digital world[J].Sustainability,2018,10(2):440. [10]威尔逊.科学研究方法论[M].石大中,译.上海:上海科学技术文献出版社,1988. [11]FLURI J,KACPRZAK T,LUCCHI A,et al.Cosmological constraints with deep learning from KiDS-450 weak lensing maps[J].Physical Review D,2019,100(6):19. [12]JUMPER J,EVANS R,PRITZEL A,et al.Highly accurate protein structure prediction with AlphaFold[J].Nature,2021,596(7873):583-589. [13]Science Official Website.2021 Breakthrough of the Year[EB/OL].(2021-12-16)[2023-06-06].https://www.science.org/content/article/breakthrough-2021#section_breakthrough. [14]BURGER B,MAFFETTONE P M,GUSEV V,et al.A mobile robotic chemist[J].Nature,2020,583(7815):237-241. [15]AHMAD S F,RAHMAT M K,MUBARIK M S,et al.Artificial intelligence and its role in education[J].Sustainability,2021,13(22):12902. [16]蔡三发,王倩,沈阳.人工智能赋能:高校学科建设的创新与发展———访中国工程院院士陈杰教授[J].电化教育研究,2020,41(2):59. [17]毛刚,王良辉.人机协同:理解并建构未来教育世界的方式[J].教育发展研究,2021,41(1):16-24. [18]CASTELVECCHI D.Deepmind AI tackles one of chemistry's top techniques[J].Nature,2021,600(371):371. [19]WANG W,SIAU K.Artificial intelligence,machine learning,automation,robotics,future of work and future of humanity:a review and research agenda[J].Journal of Database Management,2019,30(1):61-79. [20]FISCHHOFF B,DAVIS A L.Communicating scientific uncertainty[J].Proceedings of the National Academy of Sciences,2014,111(supplement_4):13664-13671. [21]SLEZAK P,BRANNIGAN A,COLLINS H M,et al.Symposium on computer discovery and the sociology of scientific knowledge[J].Social Studies of Science,1989,19(4):563-695. [22]钟义信.人工智能的范式革命与中华文明的伟大复兴[J].人民论坛·学术前沿,2023(14):29-48. [23]于水,范德志.新一代人工智能(ChatGPT)的主要特征、社会风险及其治理路径[J].大连理工大学学报(社会科学版),2023,45(4):1-7. [24]王惠敏,许峰,蔡士林.ChatGPT应用的风险迭代与法律因应研究[J].电子政务,2023(7):1-12. [25]BENDER E M,GEBRU T,MCMILLAN-MAJOR A,et al.On the dangers of stochastic parrots:can language models be too big[C]//Proceedings of the 2021 ACM Conference on Fairness,Accountability,and Transparency.2021. [26]BUOLAMWINI J,GEBRU T.Gender shades:intersectional accuracy disparities in commercial gender classification[C]//Conference on fairness,accountability and transparency.PMLR,2018. [27]RAY P P.ChatGPT:a comprehensive review on background,applications,key challenges,bias,ethics,limitations and future scope[J].Internet of Things and Cyber-Physical Systems,2023(3):121-154. [28]HARTMANN J,SCHWENZOW J,WITTE M.The political ideology of conversational AI:converging evidence on ChatGPT's pro-environmental,left-libertarian orientation[EB/OL].(2023-01-05)[2023-06-06].https://arxiv.org/abs/2301.01768. [29]ELSE H.Abstracts written by ChatGPT fool scientists[J].Nature,2023,613(7944):423-423. [30]DOŠILOVIĆ F K,BRČIĆ M,HLUPIĆ N.Explainable artificial intelligence:a survey[C]//2018 41st International Convention on Information and Communication Technology,Electronics and Microelectronics(MIPRO).IEEE,2018. [31]SCHNEIDER J,ABRAHAM R,MESKE C,et al.Artificial intelligence governance for businesses[J].Information Systems Management,2023,40(3):229-249. [32]HUTSON M.Artificial intelligence faces reproducibility crisis[J].Science,2018,359(6377):725-726. [33]GUNNING D,STEFIK M,CHOI J,et al.XAI—Explainable artificial intelligence[J].Science Robotics,2019,4(37):1-10. [34]徐英瑾,陈萌.人工智能如何“说人话”——对于自然语言处理研究的哲学反思[J].自然辩证法通讯,2022,44(1):10-19. [35]STOKEL-WALKER C.ChatGPT listed as author on research papers:many scientists disapprove[J].Nature,2023,613(7945):620-621. [36]徐英瑾.“常人”“自欺”与聊天软件[J].社会科学战线,2023,334(4):47-56. [37]ISKENDER A.Holy or unholy？Interview with open AI's ChatGPT[J].European Journal of Tourism Research,2023,34(3414):2. [38]MATTHIAS A.The responsibility gap:ascribing responsibility for the actions of learning automata[J].Ethics and Information Technology,2004,6(3):175-183. [39]郭菁.第一人称或第二人称责任——如何填补人工智能时代的责任鸿沟[J].自然辩证法研究,2023,39(4):10-16. [40]新京报.ChatGPT可以是“作者”吗[EB/OL].(2023-02-07)[2023-06-06].https://baijiahao.baidu.com/s?id=175717-8108983901298. [41]THORP H H.ChatGPT is fun,but not an author[J].Science,2023,379(6630):313. [42]EDITORIALS N.Tools such as ChatGPT threaten transparent science:here are our ground rules for their use[J].Nature,2023,613(7945):612. [43]ICML.Clarification on large language model policy LLM[EB/OL].(2023-01-17)[2023-06-06].https://icml.cc/Conferences/2023/llm-policy. [44]PIORKOWSKI D,HIND M,RICHARDS J.Quantitative ai risk assessments:opportunities and challenges[EB/OL].(2023-01-11)[2023-06-06].https://arxiv.org/abs/2209.06317. [45]MCLENNAN S,FISKE A,CELI L A,et al.An embedded ethics approach for AI development[J].Nature Machine Intelligence,2020,2(9):488-490. [46]MARJANOVIC O,CECEZ-KECMANOVIC D,VIDGEN R.Theorising algorithmic justice[J].European Journal of Information Systems,2022,31(3):269-287. [47]ZAVRŠNIK A.Algorithmic justice:algorithms and big data in criminal justice settings[J].European Journal of Criminology,2021,18(5):623-642. [48]ARNOLD M,BELLAMY R K E,HIND M,et al.Factsheets:increasing trust in AI services through supplier's declarations of conformity[EB/OL].(2019-12-07)[2023-06-06].https://arxiv.org/abs/1808.07261v2. [49]DE SILVA D,ALAHAKOON D.An artificial intelligence life cycle:from conception to production[J].Patterns,2022,3(6):100489. [50]WINFIELD A F T,JIROTKA M.Ethical governance is essential to building trust in robotics and artificial intelligence systems[J].Philosophical Transactions:Mathematical,Physical and Engineering Sciences,2018,376(2133):1-13. [51]MITTELSTADT B.Principles alone cannot guarantee ethical AI[J].Nature Machine Intelligence,2019,1(11):501-507. [52]SCHOMBERG R V.Technikfolgen abschätzen lehren:bildungspotenziale transdisziplinärer methoden[M].Wiesbaden:VS Verlag für Sozialwissenschaften,2012. [53]FELT U,WYNNE B,CALLON M.Taking European knowledge seriously[R].Brussels:European Commission,2007. [54]肖鹏.生成式人工智能对公共文化机构的影响及其负责任创新之路[J].图书馆建设,2023(4):25-28,45. [55]刘婵娟,翟渊明,刘博京.“负责任创新”的伦理内涵与实现[J].浙江社会科学,2019(3):94-99,158. [56]刘战雄,夏保华.责任过度及其对负责任创新的启示[J].自然辩证法研究,2016,32(7):41-46. [57]DIETVORST B J,SIMMONS J P,Massey C.Overcoming algorithm aversion:people will use imperfect algorithms if they can(even slightly)modify them[J].Management Science,2018,64(3):1155-1170. [58]BURTON J W,STEIN M K,JENSEN T B.A systematic review of algorithm aversion in augmented decision making[J].Journal of Behavioral Decision Making,2020,33(2):220-239. [59]SCHLEIDGEN S,FRIEDRICH O,GERLEK S,et al.The concept of“interaction”in debates on human-machine interaction[J].Humanities and Social Sciences Communications,2023,10(1):1-13.
2024, 
