HCA Healthcare says AI-driven automation of billing and clinical documentation will save roughly $400 million in 2026 — and UBS thinks the real number could reach $700 million as the technology matures across its 170+ hospitals.
A newly funded NIH project led by University of Michigan researchers is building risk-sensitive AI to prevent pharmacy dispensing errors, which contribute to an estimated 1.5 million emergency room visits annually in the US.
Insilico Medicine has advanced rentosertib, an AI-designed drug for idiopathic pulmonary fibrosis, into a Phase III trial of 320 patients, making it one of only a handful of AI-discovered therapeutics to reach late-stage testing.
Twelve major U.S. health systems serving nearly 20 million patients annually have formed the Diagnostic AI Consortium with Aidoc to jointly test and govern diagnostic AI tools, with initial results expected in 2027.
A new Chalmers University AI model dramatically speeds up molecular dynamics simulations central to drug discovery, part of a broader 2026 push by pharma companies to pool structural data and train shared AI models like OpenFold3.
Full results from the Swedish MASAI trial, involving over 105,000 women, show AI-supported mammography screening cut interval breast cancers by 12% and invasive interval cancers by 16% compared with standard double reading.
An Axios investigation found AI chatbots reliably block obvious self-harm queries but struggle to detect subtle, slow-building mental health warning signs across long conversations.
A new validation study found an AI-enhanced smartwatch algorithm detects moderate-to-severe sleep apnea with over 92% sensitivity and specificity, though it underestimates milder cases.
Texas A&M researchers unveiled an AI model trained on nearly 10,000 patients' data aimed at catching Alzheimer's disease earlier, joining a crowded 2026 field of competing early-detection AI tools.
A phishing email at healthcare AI vendor Xsolis led to a breach exposing records for 1.4 million patients across seven health systems including Mayo Clinic, underscoring the risk of concentrating sensitive data with third-party AI platforms.