Updated

ScienceDaily reported on 13 September 2026 that researchers used artificial intelligence to analyze roughly 400,000 Reddit posts written by people taking GLP-1 weight-loss and diabetes medications, including semaglutide products Ozempic and Wegovy and tirzepatide products Mounjaro and Zepbound. The analysis surfaced symptoms that patients discussed online but that are not prominent in official labeling, such as menstrual changes, chills, hot flashes and fatigue. The approach treated social media as a large-scale, real-time source of patient-reported experience. The researchers stressed that the method cannot establish that the medications caused the reported symptoms, since online posts are self-selected, unverified and lack clinical context. Even so, the work is significant because tens of millions of people now use GLP-1 receptor agonists, and traditional post-market surveillance systems capture only a fraction of side effects. The authors positioned the technique as a way to generate hypotheses that could then be tested in formal pharmacovigilance studies or trials. The study adds to a broader push to apply language models to health data.