Search papers, labs, and topics across Lattice.
This paper proposes a three-layer trust framework鈥攈uman-oriented, AI-oriented, and interaction-oriented鈥攖o reconcile differing perspectives on trustworthy AI in mental health support. The authors systematically review existing AI-driven research in mental health, analyzing evaluation practices for "trustworthiness" across automatic metrics and clinical validation. They identify critical gaps between current NLP metrics and real-world mental health needs, advocating for a socio-technically aligned research agenda.
Current NLP metrics for "trustworthy" AI in mental health are dangerously misaligned with the actual needs of patients and practitioners.
Building trustworthy AI systems for mental health support is a shared priority across stakeholders from multiple disciplines. However,"trustworthy"remains loosely defined and inconsistently operationalized. AI research often focuses on technical criteria (e.g., robustness, explainability, and safety), while therapeutic practitioners emphasize therapeutic fidelity (e.g., appropriateness, empathy, and long-term user outcomes). To bridge the fragmented landscape, we propose a three-layer trust framework, covering human-oriented, AI-oriented, and interaction-oriented trust, integrating the viewpoints of key stakeholders (e.g., practitioners, researchers, regulators). Using this framework, we systematically review existing AI-driven research in mental health domain and examine evaluation practices for ``trustworthy''ranging from automatic metrics to clinically validated approaches. We highlight critical gaps between what NLP currently measures and what real-world mental health contexts require, and outline a research agenda for building socio-technically aligned and genuinely trustworthy AI for mental health support.
