<p>The mass adoption of artificial intelligence (AI) systems, ranging from relatively simple bots to complex large language models (LLMs), poses deep challenges to the conceptual foundations of agenda setting and agenda building scholarship. AI systems blur the boundaries of what constitutes an actor in ways that researchers have yet to accommodate. In this conceptual work, we present a typology for AI systems in agenda-setting and related research which invites scholars to think of AI systems in terms of autonomy and identification. We also offer guidance for agenda-setting and related research in “the age of AI.” In particular, we suggest ways of thinking of AI systems as entities combined with human entities across different contexts that appropriately captures the dynamics of human actors using AI systems to achieve their communication goals in different contexts. Additionally, we argue for a significant extension of agenda-setting and related scholarship that prioritizes communication environments as the object of study, which we term “agenda gaming.”</p>