Position Paper Argues for Cognitive Alignment in AI Decision-Making

A new position paper on arXiv (2608.12372) argues that AI systems used in high-stakes decision-making should be cognitively aligned with their users, meaning they reason similarly to humans and clearly communicate their reasoning. The paper reviews evidence that such alignment improves understandability and trustworthiness, and presents new survey data showing many users consider cognitive alignment 'essential' when an AI's rationale is important. It outlines gaps between existing alignment methods and what is needed, and proposes a research agenda. The authors contend that cognitive misalignment is a likely barrier to AI adoption in many applications, and addressing it is crucial for creating AI systems users are willing to trust.