Knowledge Combination Analysis Reveals That Artificial Intelligence Research Is More Like “Normal Science” Than “Revolutionary Science”

Jieshu Wang, Andrew Maynard, Jose Lobo, Katina Michael, Sébastien Motsch, Deborah Strumsky

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Artificial Intelligence (AI) research is intrinsically innovative and serves as a source of innovation for research and development in a variety of domains. There is an assumption that AI can be considered “revolutionary science” rather than “normal science.” Using a dataset of nearly 300,000 AI publications, this paper examines the co-citation dynamics of AI research and investigates its trajectory from the perspective of knowledge creation as a combinatorial process. We found that while the number of AI publications grew significantly, they largely follows a normal science trajectory characterized by incremental and cumulative advancements. AI research that combines existing knowledge in highly conventional ways is a substantial driving force in AI and has the highest scientific impact. Radically new ideas are relatively rare. By offering insights into the co-citation dynamics of AI research, this work contributes to understanding its evolution and guiding future research directions.

Original languageEnglish (US)
Title of host publicationProceedings of the 57th Annual Hawaii International Conference on System Sciences, HICSS 2024
EditorsTung X. Bui
PublisherIEEE Computer Society
Pages5598-5607
Number of pages10
ISBN (Electronic)9780998133171
StatePublished - 2024
Event57th Annual Hawaii International Conference on System Sciences, HICSS 2024 - Honolulu, United States
Duration: Jan 3 2024Jan 6 2024

Publication series

NameProceedings of the Annual Hawaii International Conference on System Sciences
ISSN (Print)1530-1605

Conference

Conference57th Annual Hawaii International Conference on System Sciences, HICSS 2024
Country/TerritoryUnited States
CityHonolulu
Period1/3/241/6/24

Keywords

  • artificial intelligence
  • bibliographic data
  • co-citation networks
  • knowledge combination
  • knowledge management
  • scientific research

ASJC Scopus subject areas

  • General Engineering

Fingerprint

Dive into the research topics of 'Knowledge Combination Analysis Reveals That Artificial Intelligence Research Is More Like “Normal Science” Than “Revolutionary Science”'. Together they form a unique fingerprint.

Cite this