Artificial intelligence and the end(s) of qualitative data analysis

Large language models and other forms of artificial intelligence are increasingly harnessed in qualitative research. However, an over-reliance on AI-derived analytic tools runs the risk of normalizing thin description as a "good-enough" outcome for qualitative inquiry; bypasses the ethical arrangement between investigators and participants regarding how informants' stories and other narrative information should be handled and interpreted; and facilitates both the de-skilling of analytic practice and the alienation of workers engaging in such activities. Perhaps most important, the uncritical use of AI-based analytic tools calls into question the ontological and epistemological underpinnings of qualitative research itself.