Abstract While the possibility of reaching human-like Artificial Intelligence (AI) remains controversial, the likelihood that the future will be characterized by a society with a growing presence of autonomous machines is high. In fact, autonomous AI agents are already deployed and active across several industries and digital environments. This trajectory points to a progressive hybridization of society marked by new forms of social interaction at both micro and macro levels. Alongside traditional human-human and human-machine interactions, machine-machine interactions are poised to become increasingly prevalent. Given these developments, I argue that criminology must begin to address the implications of this transition for crime and social control. Drawing on Actor–Network Theory and Woolgar’s decades-old call for a sociology of machines – frameworks that acquire renewed relevance with the rise of AI foundation models and generative agents – I contend that criminologists should move beyond conceiving AI solely as a tool. Instead, AI agents should be recognized as entities with agency, understood as a multi-layered construct encompassing computational, social, and legal dimensions. Building on insights from the literature on AI safety, I thus examine the risks and challenges associated with the rise of multi-agent AI systems, proposing a dual taxonomy to characterize the channels through which interactions among AI agents may generate deviant, unlawful, or criminal outcomes. I then advance and discuss four key questions that warrant theoretical and empirical attention: (1) Can we assume that machines will simply mimic humans? (2) Will crime theories developed for humans hence suffice to explain deviant or criminal behaviors emerging from interactions between autonomous AI agents? (3) What types of criminal behaviors will be affected first? (4) How might this unprecedented societal shift impact policing? These questions form the core of this article, underscoring the urgent need for criminologists to theoretically and empirically engage with the implications of multi-agent AI systems for the study of crime and play a more active role in debates on AI safety and governance.

Behavioral and social scientists are skilled at diagnosing harms in our collectives – we investigate and label discrimination, exclusion, and marginalization. But how skilled are we at knowing when to stop? Identifying more harms is not necessarily better, so we maintain that the well-being of individuals and collectives requires diagnosing harm when appropriate, but also occasionally revisiting the value of those diagnoses and, when necessary, reversing those that no longer fit the data or begin to cause harm themselves. We argue that there is room for self-reflection and ask whether some persistent social diagnoses may overlook evidence of progress. To explore this possibility, we draw generously on medicine, where scholars and practitioners have long recognized that diagnostic labels can outlive their purpose. We use gender discrimination as an illustrative case to examine how well-intentioned diagnoses can drift from empirical reality and become counterproductive in some contexts. Rather than calling for uniform declassification of harm labels, we argue for diagnostic stewardship - an ongoing, empirical investigation of when diagnoses continue to prove useful and whether they begin to hinder.

This paper proposes violent capital as an analytic tool to explore violence and its relation to the social order. Following Pierre Bourdieu on cultural capital, I suggest we think of violence as a potential asset that can be invested in or exchanged for another asset, position, or status. After an introductory discussion of conceptualization and operationalization, and to bring the concept to life and suggest analytic utility, the paper turns to violent capital and hierarchies. Given the centrality of states to the history of violence (Tilly 1992; Mann 1986, 1993; North, Wallis, and Weingast 2009), examining violent capital and hierarchies provides further insights into the nature of states and state-building, and vice versa. The paper uses historical cases to show three different ways of the relation of violent capital to status, that also expand our understanding of the rise of the state: antiquity (Sparta and Rome), medieval Europe (knights), and modern armies. These cases reveal variation in how violent capital is invested and exchanged and in its value for hierarchies. These cases also suggest additional dimensions to the historical evolution of the modern state and “coercion” as more than resources for physical force: this evolution is also disciplining, compartmentalizing, and subordinating violent capital. The paper concludes with a research agenda and further potential analytic utility.