Paper Type
Complete
Abstract
As AI systems increasingly participate in team-based work, understanding how they function as teammates has become a critical research concern. This study presents a systematic literature review of 37 studies on human–AI collaboration, integrating three analytical dimensions: AI role positioning (tool vs. partner), interaction structure (one-to-one, one-to-many, many-to-one, many-to-many), and affordance manifestation (cognitive and affective). Following PRISMA guidelines, we develop a concept matrix to systematically map how AI affordances emerge across varying collaborative configurations. Our findings indicate that cognitive affordances are present across both role positions, while affective affordances are more prominent when AI is positioned as a partner in distributed settings. We further demonstrate that affordance manifestation is structurally contingent rather than purely capability-driven. This study offers an integrative framework for advancing theory on human–AI teamwork and proposes a structured research agenda for future empirical investigation.
Paper Number
1704
Recommended Citation
Yeh, Gary Yu-Ho; Cheng, Yuting; Hsu, Jack; Cheng, Hsiang-Lan of Long-Term; and Chiu, Chao-Min, "AI as Teammate in Human–AI Collaboration: A Literature Review of Interaction Structures and Affordances" (2026). AMCIS 2026 Proceedings. 10.
https://aisel.aisnet.org/amcis2026/sigcnow/sigcnow/10
AI as Teammate in Human–AI Collaboration: A Literature Review of Interaction Structures and Affordances
As AI systems increasingly participate in team-based work, understanding how they function as teammates has become a critical research concern. This study presents a systematic literature review of 37 studies on human–AI collaboration, integrating three analytical dimensions: AI role positioning (tool vs. partner), interaction structure (one-to-one, one-to-many, many-to-one, many-to-many), and affordance manifestation (cognitive and affective). Following PRISMA guidelines, we develop a concept matrix to systematically map how AI affordances emerge across varying collaborative configurations. Our findings indicate that cognitive affordances are present across both role positions, while affective affordances are more prominent when AI is positioned as a partner in distributed settings. We further demonstrate that affordance manifestation is structurally contingent rather than purely capability-driven. This study offers an integrative framework for advancing theory on human–AI teamwork and proposes a structured research agenda for future empirical investigation.
When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.
Comments
SIG CNOW