Reflexive Human–AI Collaboration: Tracing the Evolving Epistemics of Qualitative Inquiry (2021–2025)

This study maps the evolution of Artificial Intelligence (AI) in qualitative research from 2021 to 2025 using SciMAT analyses of 1,862 Scopus-indexed publications. Three phases emerged: (1) early experimentation and methodological hybridization (2021–2023), (2) systematization and applied integration (2024), and (3) institutional maturity and ethical rebalancing (2025). In the first phase, AI entered qualitative inquiry through assistive functions, mainly transcription, coding support, and sentiment analysis, primarily in health and social science research. Themes such as Semi-Structured Interview and Qualitative Research anchored this stage, reflecting efforts to merge computational efficiency with interpretive depth. By 2024, AI methods became routine in qualitative workflows. Clusters including Interview , Patient Care , and ChatGPT show how NLP and large language models supported transcript analysis, coding, and focus-group simulation while prompting debates on reliability, validity, and human interpretive control. By 2025, the field exhibited institutional consolidation. Major themes, such as Health Personnel Attitude , Students , Human , and Qualitative Analysis , signaled the rise of ethical governance, AI literacy in graduate training, and increased attention to equity and contextual sensitivity. AI was increasingly viewed as a reflexively managed collaborator rather than a replacement for human analysis. The findings reveal a clear trajectory from early hybrid experimentation to reflexive human–AI partnership. The study demonstrates how qualitative research is being reorganized technically, ethically, and pedagogically, and highlights the principles required to ensure that AI-enhanced inquiry remains human-centered and interpretively robust.