The Reasons that Agents Act: Intention and Instrumental Goals
This work addresses the foundational problem of understanding and formalizing intention in AI, which is crucial for concepts like agency and safety, but it is incremental as it builds on existing philosophical and causal frameworks.
The paper tackles the challenge of ascribing intention to AI systems by introducing a formal definition of intention in structural causal influence models, grounded in philosophy, and demonstrates its application to infer intentions from the behavior of reinforcement learning agents and language models.
Intention is an important and challenging concept in AI. It is important because it underlies many other concepts we care about, such as agency, manipulation, legal responsibility, and blame. However, ascribing intent to AI systems is contentious, and there is no universally accepted theory of intention applicable to AI agents. We operationalise the intention with which an agent acts, relating to the reasons it chooses its decision. We introduce a formal definition of intention in structural causal influence models, grounded in the philosophy literature on intent and applicable to real-world machine learning systems. Through a number of examples and results, we show that our definition captures the intuitive notion of intent and satisfies desiderata set-out by past work. In addition, we show how our definition relates to past concepts, including actual causality, and the notion of instrumental goals, which is a core idea in the literature on safe AI agents. Finally, we demonstrate how our definition can be used to infer the intentions of reinforcement learning agents and language models from their behaviour.