A Causal Model of Theory of Mind in Conflict for Artificial Intelligence

arXiv cs.AI · 18 hours ago

Researchers developed a causal model that determines when artificial intelligence systems should engage in theory of mind reasoning during conflict situations, rather than always activating this capacity. The model uses a directed acyclic graph with four exogenous variables and three causal pathways to decide whether mentalizing improves epistemic accuracy. The framework aims to make AI systems more efficient and trustworthy in human-machine teaming by providing a resource-rational decision procedure for social reasoning.