Normative Equivalence in Human-AI Cooperation: Behaviour, Not Identity, Drives Cooperation in Mixed-Agent Groups

TL;DR

This study investigates whether introducing an AI-labelled teammate alters cooperative norms in small groups using a $PGG$ with 3 humans and 1 bot across label ($human$ vs $AI$) and strategy ($Unconditional$, $Conditional$, $Free-Rider$) manipulations. It combines ten rounds of a repeated $PGG$, a follow-up one-shot $PD$, and normative perception measures, with a total sample of $N=236$ participants. The results show that reciprocal group dynamics and behavioural inertia predominantly shape cooperation, and these normative processes operate identically regardless of the bot’s label, providing evidence for a form of normative equivalence between mixed human–AI and all-human groups. Practically, the findings suggest cooperative norms can extend to artificial agents in settings with minimal social presence, informing AI design in teams and highlighting directions for testing adaptive or more socially prominent AI in future work.

Abstract

The introduction of artificial intelligence (AI) agents into human group settings raises essential questions about how these novel participants influence cooperative social norms. While previous studies on human-AI cooperation have primarily focused on dyadic interactions, little is known about how integrating AI agents affects the emergence and maintenance of cooperative norms in small groups. This study addresses this gap through an online experiment using a repeated four-player Public Goods Game (PGG). Each group consisted of three human participants and one bot, which was framed either as human or AI and followed one of three predefined decision strategies: unconditional cooperation, conditional cooperation, or free-riding. In our sample of 236 participants, we found that reciprocal group dynamics and behavioural inertia primarily drove cooperation. These normative mechanisms operated identically across conditions, resulting in cooperation levels that did not differ significantly between human and AI labels. Furthermore, we found no evidence of differences in norm persistence in a follow-up Prisoner's Dilemma, or in participants' normative perceptions. Participants' behaviour followed the same normative logic across human and AI conditions, indicating that cooperation depended on group behaviour rather than partner identity. This supports a pattern of normative equivalence, in which the mechanisms that sustain cooperation function similarly in mixed human-AI and all human groups. These findings suggest that cooperative norms are flexible enough to extend to artificial agents, blurring the boundary between humans and AI in collective decision-making.

Figures (9)

• Figure 1: Schematic representation of the experimental design and PGG mechanics
• Figure 2: Mean Contribution per Round by Treatment and Bot Strategy.
• Figure 3: Prisoner's Dilemma choices by treatment and bot strategy. The first letter denotes the prediction of the partner's behaviour, and the second letter denotes the participant's choice.
• Figure 4: Prisoner's Dilemma choices by treatment and bot strategy.
• Figure 5: Predicted probability of PD cooperation by prior PGG contributions at the (a) individual and (b) group level.