Towards Directive Explanations: Crafting Explainable AI Systems for Actionable Human-AI Interactions

TL;DR

The paper tackles the need for actionable explanations for non-expert users in AI-driven decision support and argues that static explanations are insufficient for guiding behavior. It advances directive explanations and interactive, data-centric approaches implemented through healthcare-focused studies, including the Explanatory Model Steering (EXMOS) framework and an explainability dashboard. Key findings show that combining multiple explanation types and granting domain experts greater control over data-centric disclosures improves understandability, trust, and actionability, with practical design guidelines for healthcare XAI. By proposing LLM-based conversational explanations and objective evaluation metrics, the work lays groundwork for more effective human-AI collaboration in real-world settings.

Abstract

With Artificial Intelligence (AI) becoming ubiquitous in every application domain, the need for explanations is paramount to enhance transparency and trust among non-technical users. Despite the potential shown by Explainable AI (XAI) for enhancing understanding of complex AI systems, most XAI methods are designed for technical AI experts rather than non-technical consumers. Consequently, such explanations are overwhelmingly complex and seldom guide users in achieving their desired predicted outcomes. This paper presents ongoing research for crafting XAI systems tailored to guide users in achieving desired outcomes through improved human-AI interactions. This paper highlights the research objectives and methods, key takeaways and implications learned from user studies. It outlines open questions and challenges for enhanced human-AI collaboration, which the author aims to address in future work.

Figures (1)

• Figure 1: Overview of the current research progress and planned next steps to address research questions