Social agents with finitely nested opponent models are vulnerable to manipulation by agents with deeper recursive capabilities. This imbalance, rooted in logic and the theory of recursive modelling frameworks, cannot be solved directly. We propose a computational framework called ℵ-IPOMDP, which augments the Bayesian inference of model-based RL agents with an anomaly detection algorithm and an out-of-belief policy. Our mechanism allows agents to realize that they are being deceived, even if they cannot understand how, and to deter opponents via a credible threat. We test this framework in both a mixed-motive and a zero-sum game. Our results demonstrate the ℵ-mechanism’s effectiveness, leading to more equitable outcomes and less exploitation by more sophisticated agents. We discuss implications for AI safety, cybersecurity, cognitive science, and psychiatry.

Effective communication in Multi-Agent Reinforcement Learning (MARL) can significantly enhance coordination and collaborative performance in complex and partially observable environments. However, reliance on communication can also introduce vulnerabilities when agents are misaligned, potentially leading to adversarial interactions that exploit implicit assumptions of cooperative intent. Prior work has addressed adversarial behavior through power regularization through controlling the influence one agent exerts over another, but has largely overlooked the role of communication in these dynamics. This paper introduces Communicative Power Regularization (CPR), extending power regularization specifically to communication channels. By explicitly quantifying and constraining agents’ communicative influence during training, CPR actively mitigates vulnerabilities arising from misaligned or adversarial communications. Evaluations across benchmark environments Red-Door-Blue-Door, Predator-Prey, and Grid Coverage demonstrate that our approach significantly enhances robustness to adversarial communication while preserving cooperative performance, offering a practical framework for secure and resilient cooperative MARL systems.

Large language models (LLMs) can provide users with false, inaccurate, or misleading information, and we consider the output of this type of information as what Natale calls ‘banal’ deceptive behaviour [53]. Here, we investigate peoples’ perceptions of ChatGPT-generated deceptive behaviour and how this affects people’s behaviour and trust. To do this, we use a mixed-methods approach comprising of (i) an online survey with 220 participants and (ii) semi-structured interviews with 12 participants. Our results show that (i) the most common types of deceptive information encountered were over-simplifications and outdated information; (ii) humans’ perceptions of trust and chat-worthiness of ChatGPT are impacted by ‘banal’ deceptive behaviour; (iii) the perceived responsibility for deception is influenced by education level and the perceived frequency of deceptive information; and (iv) users become more cautious after encountering deceptive information, but they come to trust the technology more when they identify advantages of using it. Our findings contribute to understanding human-AI interaction dynamics in the context of Deceptive AI Ecosystems and highlight the importance of user-centric approaches to mitigating the potential harms of deceptive AI technologies.

Online reviews now have a considerable influence on consumer choices. However, little work has focused on what features of review platforms influence review quality. We present a novel approach to identify the features that encourage quality reviews. By interpreting reviews as arguments for or against the product, an argument scheme can be used to simulate the emergent reliability of reviews resulting from different setups of the online review platform. Our results show that if the most recent, helpful, or polarised reviews are promoted over quality, then good quality reviews will almost never be shown to users.

Large Language Models (LLMs) have become a significant milestone in the history of artificial intelligence, representing a powerful technology that drives advancements in natural language understanding and generation. In this paper, we propose an approach in which LLMs are utilized to support the task of translating natural language arguments into computational representations. Our approach is grounded in using argumentation schemes to classify arguments, providing context to LLMs for performing the proposed task. Our results demonstrate that LLMs, even with a short context, can handle simple argument structures. Moreover, our findings suggest that a larger context would likely enhance the performance, particularly when dealing with more complex argument structures.

Self-governing hybrid societies are multi-agent systems where humans and machines interact by adapting to each other’s behaviour. Advancements in Artificial Intelligence (AI) have brought an increasing hybridisation of our societies, where one particular type of behaviour has become more and more prevalent, namely deception. Deceptive behaviour as the propagation of disinformation can have negative effects on a society’s ability to govern itself. However, self-governing societies have the ability to respond to various phenomena. In this article, we explore how they respond to the phenomenon of deception from an evolutionary perspective considering that agents have limited adaptation skills. Will hybrid societies fail to govern deceptive behaviour and reach a Tragedy of The Digital Commons? Or will they manage to avoid it through cooperation? How resilient are they against large-scale deceptive attacks? We provide a tentative answer to some of these questions through the lens of evolutionary agent-based modelling, based on the scientific literature on deceptive AI and public goods games.

Personal assistants (PAs) such as Amazon Alexa, Google Assistant and Apple Siri are now widespread. However, without adequate safeguards and controls their use may lead to privacy risks and violations. In this paper, we propose a model for privacy-enhancing PAs. The model is an interpretable AI architecture that combines 1) a dialogue mechanism for understanding the user and getting online feedback from them, with 2) a decision-making mechanism based on case-based reasoning considering both user and scenario similarity. We evaluate our model using real data about users’ privacy preferences, and compare its accuracy and demand for user involvement with both online machine learning and other, more interpretable, AI approaches. Our results show that our proposed architecture is more accurate and requires less intervention from the users than existing approaches.

Smart Personal Assistants (SPAs), such as Amazon Alexa, Google Assistant and Apple Siri, leverage different AI techniques to provide convenient help and assistance to users. However, inappropriate information sharing decisions can lead SPAs to incorrectly disclose user information to undesired parties, or mistakenly block their reasonable access in specific scenarios to desired parties. In fact, reports about privacy violations in SPAs and associated user concerns are well known and understood in the related literature. It is difficult for SPAs to automatically decide how data should be shared with respect to the privacy preferences of the users. We argue norms, which are regarded as shared standards of acceptable behaviour of groups and/or individuals, can be used to govern and reason about the best course of action of SPAs with regards to information sharing, and our work is the first to propose a practical model to address the above issues and govern SPAs based on normative systems and the contextual integrity theory of privacy. We evaluated the performance of the model using a real dataset of user preferences for privacy in SPAs and the results showed a very marked and significant improvement in understanding user preferences and making the right decisions with respect to data sharing.

Recent events that revolve around fake news indicate that humans are more susceptible than ever to mental manipulation by powerful technological tools. In the future these tools may become autonomous. One crucial property of autonomous agents is their potential ability to deceive. From this research we hope to understand the potential risks and benefits of deceptive artificial agents. The method we propose to study deceptive agents is by making them interact with agents that detect deception and analyse what emerges from these interactions given multiple setups such as formalisations of scenarios inspired from historical cases of deception.