The Role of Artificial Intelligence in Transforming Psychological Help-Seeking Behaviors: A Social–Behavioral Analysis of User Trust, Acceptance, and Resistance

Abstract

This study aimed to explain how artificial intelligence transforms psychological help-seeking behaviors and to analyze the social and behavioral processes underlying users’ trust, acceptance, and resistance. This qualitative study employed a reflexive thematic analysis approach. Participants were 26 residents of Tehran who had used generative artificial intelligence systems or conversational chatbots at least three times during the previous six months to obtain mental health information, seek emotional support, or decide whether to consult a mental health professional. Purposive maximum-variation sampling was used to ensure diversity in gender, age, educational level, occupational status, and experience with artificial intelligence. Data were collected exclusively through individual semistructured interviews. After obtaining informed consent, the interviews were audio-recorded, transcribed verbatim, and organized and analyzed using NVivo 14. Theoretical saturation was reached during the twenty-third interview, followed by three additional interviews to confirm informational adequacy. Four main categories were identified: “low-friction access and a reduced threshold for help-seeking,” “conditional and evaluative trust,” “instrumental acceptance and preference for a human–AI hybrid model,” and “social–emotional resistance and protection of the uniqueness of human relationships.” Participants perceived continuous availability, anonymity, affordability, immediate responses, and the opportunity to disclose embarrassing concerns without fear of judgment as important facilitators of help-seeking. Nevertheless, trust was provisional and was repeatedly reassessed according to response accuracy, transparency, cultural relevance, privacy protection, consistency, and the system’s ability to recognize its limitations. Most participants accepted artificial intelligence as a resource for psychoeducation, emotional clarification, preliminary support, symptom monitoring, and preparation for professional consultation, but rejected its use as a complete substitute for therapists. Artificial intelligence appears to transform psychological help-seeking from a binary decision to seek or avoid professional care into a gradual, experimental, and multi-resource process. User acceptance does not necessarily indicate complete trust; rather, it reflects a continuing negotiation among convenience, perceived usefulness, vulnerability, risk, and the need for human understanding. Responsible integration of artificial intelligence into mental health services therefore requires transparent communication, culturally and linguistically appropriate design, strong data protection, professional oversight, crisis safeguards, and explicit referral pathways to qualified human professionals.