How universities become ready for AI-supported crisis intervention: a hierarchical model of psychological and institutional readiness

The increasing use of artificial intelligence (AI) in higher education has expanded possibilities for data-informed psychological crisis intervention. Grounded in the Technological Pedagogical Readiness (TPR) framework and crisis-intervention theory, this study develops and validates a hierarchical model of perceived AI-supported crisis-intervention readiness. Here, perceived readiness refers to university staff members’ self-reported preparedness to interpret AI-generated information and make ethically responsible crisis-intervention judgments. Using survey data from 906 university staff members in China, partial least squares structural equation modeling was employed to test a reflective–reflective second-order model linking external (professional development, contextual conditions, student technological literacy, parent/community involvement) and internal (AI-TPACK, perceived usefulness, perceived ease of use, ethical competence) readiness factors to three domains of crisis readiness: developmental, situational, and existential. Results indicate that perceived crisis-intervention readiness is multi-layered and domain-specific rather than unitary. External readiness was positively associated with all crisis domains and substantial indirect effects through internal readiness, supporting a statistical dual-pathway pattern. Internal readiness emerged as the core psychological correlate of AI-supported judgment and ethical decision-making. The findings extend TPR research beyond instructional contexts into mental-health governance and position AI as supporting, rather than replacing, professional judgment in higher education crisis intervention.