Beyond Injection Detection: A Positive-Security Prompt Firewall that Closes the Scope and PHI Gap SOTA Classifiers Miss in Healthcare
June 12, 2026Standard security tools fail to stop medical AI from leaking patient data because the most dangerous clinical hacks look like completely normal requests.
AI Diagnoses Brain Tumors in Twelve Minutes
June 11, 2026A new computational pathology model exposes the limits of human sight in oncology, diagnosing complex brain tumors in minutes rather than weeks.
Simple Math Beats LLMs in ICU Shock Prediction
June 11, 2026A new study reveals that basic physiological math outpaces complex large language models at predicting patient crash times.
AI diagnoses kidney disease with fewer labeled scans
June 11, 2026A new active learning model proves that medical AI does not need massive, expensive datasets to outperform human-labeled benchmarks.
New benchmark exposes AI medication safety flaws
June 11, 2026Hospital IT departments eyeing generative AI for medication safety may want to pause, as a new benchmark reveals that some top-tier models behave like hyper-anxious assistants that flag almost everything as a hazard.
Brain scans predict teen depression one year early
June 10, 2026A new computational model uses a single brain scan to predict adolescent anxiety and depression before symptoms fully emerge.
Replacing Doctors With Code
June 10, 2026The federal push to let autonomous AI diagnose patients and prescribe drugs is setting up a dangerous collision between political speed and medical safety.
AI predicts severe drug-induced liver injury without biopsies
June 10, 2026A new machine learning model uses dual ultrasound measurements to identify severe liver damage, offering a way to bypass painful and risky biopsies.
EMA Accepts AI Clinical Trial Evidence
June 10, 2026Regulators just crossed a digital rubicon by accepting AI-generated pathology data to evaluate metabolic liver disease treatments.