How is AI shaping the sensory practices and ethical sensibilities of contemporary American mental healthcare? My current book project explores this question through an ethnographic investigation of efforts to embed voice analysis technologies, especially “vocal biomarker AI,” into American mental healthcare research and practice in an attempt transform how people who interface with the mental healthcare system are listened to.

Speech, Signal, Symptom (under contract, University of Chicago Press) maps the complex and contradictory ways that machine listening technologies retrench, rather than reorient, the American mental healthcare system’s dominant interpretive paradigms and thresholds of viability. At the same time, it turns up the volume on the self-reflexive critiques and strategic maneuverings that people engaged in constructing these technologies—from data annotators to human research subjects—develop through their perspective as the ones most responsible for iteratively transducing speech into computationally tractable and clinically valuable signals. In so doing, the book articulates the stakes and limitations of framing America’s mental health crisis as a computational problem that can be resolved through techno-sensory means alone.

Research from this project was funded by the National Science Foundation, the Wenner-Gren Foundation, and the Society for Psychological Anthropology. It has been published in Science, Technology, & Human Values, Somatosphere, the AI Now Institute’s A New Lexicon of AI series, the edited volume Technocreep and the Politics of Things Not Seen (May 2025), and will appear in a forthcoming issue of Current Anthropology.