A sweeping review found that most AI-enabled medical devices authorized for patient care by the FDA had not been tested for patient outcomes. Researchers identified 1,357 AI-enabled devices authorized by the agency through December 5, 2025 and reported that only a small fraction evaluated whether the tools improved health outcomes in clinical settings. The review highlighted a gap between regulatory clearance—often based on performance metrics—and proof that the devices deliver measurable benefit for patients, emphasizing the need for stronger evidence standards as AI deployment expands across care pathways. The findings underscore the tension in medical AI regulation between technical validation and clinical effectiveness, particularly for devices that are intended to influence diagnosis, triage, or treatment selection.