Causal Foundations of Collective Agency

TL;DR

This paper introduces a causal framework to determine when groups of agents can be considered as unified collective agents, aiding understanding and control of emergent behaviors in multi-agent AI systems.

Contribution

It formalizes collective agency using causal games and abstraction, providing a new behavioral perspective and tools for analyzing multi-agent interactions.

Findings

Solved a multi-agent incentive puzzle in actor-critic models.

Quantitatively assessed collective agency in voting mechanisms.

Provided a foundation for predicting emergent collective behaviors.

Abstract

A key challenge for the safety of advanced AI systems is the possibility that multiple simpler agents might inadvertently form a collective agent with capabilities and goals distinct from those of any individual. More generally, determining when a group of agents can be viewed as a unified collective agent is a foundational question in the study of interactions and incentives in both biological and artificial systems. We adopt a behavioral perspective in answering this question, ascribing collective agency to a group when viewing the group's joint actions as rational and goal-directed successfully predicts its behavior. We formalize this perspective on collective agency using causal games -- which are causal models of strategic, multi-agent interactions -- and causal abstraction -- which formalizes when a simple, high-level model faithfully captures a more complex, low-level model. We use this framework to solve a puzzle regarding multi-agent incentives in actor-critic models and to make quantitative assessments of the degree of collective agency exhibited by different voting mechanisms. Our framework aims to provide a foundation for theoretical and empirical work to understand, predict, and control emergent collective agents in multi-agent AI systems.