JAMA+ AI

Announcing the launch of JAMA+ AI, an engaging, interactive channel that amplifies the best of the JAMA Network’s content exploring the science of artificial intelligence and digital medicine and its application in health and health care. ja.ma/3By7x4t

RCT: For adults with generalized #anxiety disorder, smartphone-delivered digital #cognitive behavioral therapy yielded greater anxiety reduction and remission rates at posttreatment follow-up vs psychoeducation. ja.ma/4uAoQsS

AI-powered #diabetic retinopathy screening at FQHCs may boost screening rates, streamline referrals, and improve care access for underserved patients. ja.ma/4bi9kKI

💬 Viewpoint: Preemption of state medical #AI laws by executive order challenges constitutional limits and risks undermining innovation and patient protections; congressional action is required for federal policy. ja.ma/4bOdaeA

Among simulated gastroenterology cases, recommendations from large language models (#LLMs) varied according to patient sociodemographic factors. ja.ma/40VSqeV

Yulin Hswen, ScD, MPH, Associate Editor of JAMA and JAMA+ AI, will be presenting at the #ASPPH2026 Annual Meeting on Friday, March 20, 2026, from 9:15 AM to 10:15 AM ET. Learn more about the presentation here: ja.ma/3N8NNLf

This study of consumer #chatbots prompted with adolescent health emergencies found that companion chatbots featured fewer safeguards than general-assistant chatbots, with consistently poor performance in crisis management. ja.ma/4uq4mCW

Pilot RCT: Children wore a #smartwatch to anticipate tantrums 75% of the time and parents responded to behavior prompts for proactive #parenting skills in <4 seconds, suggesting smartwatch-augmented parent-child interaction therapy warrants further study. jamanetwork.com/journals/jaman…

Robert Wachter, MD, spoke with JAMA+ AI Editor-in-Chief Roy Perlis, MD, MSc, about how #GenerativeAI tools are already ubiquitous among trainees and clinicians and what's next for digital platforms in #medicine. 🎧 Listen now: ja.ma/3PhvwvK

Last week, Yulin Hswen, ScD, MPH, Associate Editor at JAMA and JAMA+ AI, presented at @theNASEM workshop on Policy Issues for Integrating #ArtificialIntelligence in #Cancer Research and Care. 🎥 Watch the full workshop recording here: ja.ma/4dlcSNu

In this episode of JAMA+ AI Conversations, JAMA+ AI Editor in Chief Roy Perlis, MD, MSc, interviews OpenAI’s ChatGPT 🤖 (GPT-4o, voice mode) about the development and legacy of the first clinical chatbots, ELIZA and PARRY. 🎧 Listen now: ja.ma/4bfCpoF

#MachineLearning analyses in the MYRIAD trial found only clinically trivial personalization of school-based mindfulness for adolescent #depression prevention, underscoring limited differential benefit in universal programs. jamanetwork.com/journals/jamap…

Narrative review: Large language models (#LLMs) can structure clinical notes, automate phenotyping, and augment decision support in otolaryngology but require validation beyond feasibility and secure, open-source deployment. ja.ma/4sG2aFZ

In this episode, @JAMAplusAI Associate Editor Yulin Hswen, ScD, MPH, speaks with David Wu, MD, PhD, and Adam Rodman, MD, MPH, about what safe clinical use of #LLMs requires. 🎧 Listen now: ja.ma/476RgjY

#AI technologies hold potential to enhance acute myeloid leukemia management by improving risk stratification, early detection, and treatment optimization through integration of clinical, morphological, and genetic data. ja.ma/4aKI0Ex

This Review summarizes recent #DeepLearning advances in otolaryngology, including diagnostic models with expert-level accuracy for nasopharyngeal carcinoma and otologic pathology. ja.ma/4u5OMwa

Artificial intelligence tools may improve access and quality in #MentalHealth care, but implementation poses risks including diminished clinician skill and unintended consequences. ja.ma/4rNuVAo

High exposure to adverse childhood or adulthood experiences was associated with increased #dementia and #stroke risk in Chinese adults, partly mediated by depression, with highest risk in those exposed during both life stages. jamanetwork.com/journals/jaman…

Nearly a third of US youth aged 4 to 17 used #GenAI applications, with adoption rates exceeding 50% among teens, and a subset engaged in heavy daily use. ja.ma/4spjkaE

A vendor-agnostic #deeplearning model identified multiple #MacularDisease categories and provided triage recommendations from 3D optical coherence tomography (OCT) scans, achieving high accuracy for urgent and semi-urgent cases across OCT vendors. ja.ma/4coWB9P

#AI-powered #diabetic retinopathy screening at FQHCs may boost screening rates, streamline referrals, and improve care access for underserved patients. jamanetwork.com/journals/jaman…