Investigating hybridity in artificial intelligence research

Research in the global field of artificial intelligence is increasingly hybrid orientation. Researchers are beholden to requirements multiple intersecting spheres, such as scholarly, public, and commercial, each with their own language logic. Relatedly, collaboration across disciplinary, sector national borders expected, or required. Using a dataset 93,482 publications, this article operationalises commercial spheres through citations, news mentions, patent respectively. High performing publications (99th percentile) for metric were separated into eight categories influence. These comprised four blended influence (news, patents citations; patents; citations) three single (citations; news; patents), addition ‘Other’ category non-high publications. The develops applies two components new hybridity lens: evaluative generative hybridity. multinomial logistic regression, selected aspects knowledge production – research context, focus, artefacts, collaborative configurations examined. results elucidate key characteristics demonstrate utility proposed lens.