4 artifacts matching
Peer-reviewed
Generative Artificial Intelligence and Maternal Health Outcomes in Pregnant and Postpartum Individuals: A Randomized Clinical Trial
Two-group randomized clinical trial of access to a generative AI maternal health chatbot versus an active control (monthly children's books) among pregnant or parenting people from racial and ethnic…
Peer-reviewed
Silent Manipulation of Mental Health Treatment Recommendations From a Large Language Model
Quality-improvement study of whether inference-time activation steering, with no change to model weights or the prompt, alters an open-weights language model's answers about treatment for depression.…
Peer-reviewed
Prevalence of Mental Health Discussions in Publicly Available Generative AI Conversations
Cross-sectional research letter analysing 620,699 ChatGPT conversations from the public WildChat-4.8M corpus with an LLM-based classifier that scores topicality, intent, clinical language and affecti…
Peer-reviewed
Generative AI Use and Depressive Symptoms Among US Adults
Cross-sectional survey study examining the association between the extent and type of generative AI use and depressive symptoms (PHQ-9) among 20,847 US adults drawn from a 50-state nonprobability int…