Socially-Minded Intelligence: How Individuals, Groups, and AI Systems Can Make Each-Other Smarter (or Not) (2409.15336v2)

Abstract: A core part of human intelligence is the ability to work flexibly with others to achieve both individual and collective goals. The incorporation of artificial agents into human spaces is making increasing demands on AI to demonstrate and facilitate this ability. However, this kind of flexibility is not well understood because existing approaches to intelligence typically focus either on the individual or the collective level of analysis. At the individual level, intelligence is seen as an individual-difference trait that exists independently of the social environment. At the collective level intelligence is conceptualized as a property of groups, but not in a way that can be used to understand how groups can make group members smarter or how group members acting as individuals might make the group itself more intelligent. In the present paper we argue that by focusing either on individual or collective intelligence without considering their interaction, existing conceptualizations of intelligence limit the potential of people and machines. To address this impasse, we identify and explore a new kind of intelligence - socially-minded intelligence - that can be applied to both individuals (in a social context) and collectives (of individual minds). From a socially-minded intelligence perspective, the potential intelligence of individuals is unlocked in groups, while the potential intelligence of groups is maximized by the flexible, context-sensitive commitment of individual group members. We propose ways in which socially-minded intelligence might be measured and cultivated within people, as well as how it might be modelled in AI systems. Finally, we discuss ways in which socially-minded intelligence might be used to improve human-AI teaming.