Algorithmic Collective Action with Multiple Collectives
This work addresses the challenge of coordinating user-side steering in AI systems for broader societal impact, representing an incremental advancement by extending ACA from single to multiple collectives.
The paper tackles the problem of Algorithmic Collective Action (ACA) with multiple collectives, which are decentralized groups aiming to steer learning systems by altering shared data, by developing the first theoretical framework for such settings and providing quantitative results on the interplay of collective sizes and goal alignment.
As learning systems increasingly influence everyday decisions, user-side steering via Algorithmic Collective Action (ACA)-coordinated changes to shared data-offers a complement to regulator-side policy and firm-side model design. Although real-world actions have been traditionally decentralized and fragmented into multiple collectives despite sharing overarching objectives-with each collective differing in size, strategy, and actionable goals, most of the ACA literature focused on single collective settings. In this work, we present the first theoretical framework for ACA with multiple collectives acting on the same system. In particular, we focus on collective action in classification, studying how multiple collectives can plant signals, i.e., bias a classifier to learn an association between an altered version of the features and a chosen, possibly overlapping, set of target classes. We provide quantitative results about the role and the interplay of collectives' sizes and their alignment of goals. Our framework, by also complementing previous empirical results, opens a path for a holistic treatment of ACA with multiple collectives.