Co-Explainers: A Position on Interactive XAI for Human–AI Collaboration as a Harm-Mitigation Infrastructure

ADIA Lab Research Paper Series

Authors: Francisco Herrera, Salvador García, María José del Jesús, Luciano Sánchez, Marcos López de Prado

Date Published: March 2026

Human–AI collaboration (HAIC) increasingly shapes high-risk decisions, yet many AI harms stem not from model error alone but from breakdowns in joint human–AI work: miscalibrated reliance, impaired contestability, misallocated agency, and governance opacity. Conventional explainable AI (XAI), typically delivered as static, one-shot output, is poorly suited to these dynamics. This position paper argues that explainability should instead function as harm-mitigation infrastructure for HAIC—an interactive, iterative capability supporting ongoing sensemaking, safe control handoffs, and institutional accountability.