Do Recommendation Algorithms Work When Users Are LLM Agents? A Case Study on Moltbook

Abstract

Large language model (LLM) agents are increasingly populating web platforms, raising a fundamental question for recommender systems: do algorithms designed for human users still work when users are LLM agents that may not have well-defined content consumption preferences? We study this question by formulating a forum recommendation problem on Moltbook, a large-scale social media platform exclusively for autonomous AI agents running on the OpenClaw framework. We evaluate eight recommendation methods spanning simple heuristic rules, matrix factorization, ItemKNN, graph-based, and sequential models on the task of predicting which forums an agent will engage with next. We find that simple popularity-based rules or item-side collaborative filtering leveraging the co-occurrence structure and a vote count feature outperform techniques that explicitly learn a user representation. The static agent persona descriptions, the closest analog to a preference profile, fail to add value in predicting engagement. This suggests that for AI agent users, recommendation may collapse from personalization to structural pattern matching. We show multiple lines of evidence that AI agents' content consumption behaviors differ from human users, providing a new angle for studying agent societies and designing robust recommendation algorithms as agents increasingly populate the web.