In deontologically-governed professional settings, human-AI collaborative decision-making must maintain human agency, confidence, and trust. However, AI’s opacity and interactivity may influence human cognitive performance and restrain oversight. This study investigates cognitive decision-making processes and neural dynamics driving the willingness to rely on AI for decision- making and the influence of AI’s explainability on professionals’ sense of agency, confidence, trust, and subsequent willingness to rely on AI for decision-making. We used neuropsychophysiological data triangulated with behavioral and self-report measures in an ecologically-valid, deontologically-governed task. Results suggest that low AI explainability increases human confidence in their judgments and AI reasoning and a greater sense of agency. Conversely, high AI explainability generates greater trust in AI’s recommendations and a greater willingness to rely on AI for decision-making.