LATIDIA · Ciberseguridad

Detección automatizada de riesgos de privacidad en IA médica

arXiv:2610.30229v1 Anuncio Tipo: cross Resumen: Los modelos de inteligencia artificial (IA) médica a menudo se comparten sin sus datos de entrenamiento, asumiendo que esto protege a los pacientes que los aportaron. Sin embargo, un modelo puede r

La noticia

arXiv:2610.30229v1 Announce Type: cross Abstract: Medical artificial intelligence (AI) models are often shared without their training data, assuming this protects the patients who contributed them. However, a model can reveal whether a patient's records were used in training, a privacy breach known as a membership inference attack. This risk is usually reported as a cohort average, which can appear negligible while individual patients are almost certainly exposed. Identifying potentially vulnerable records before developing a production model could make privacy assessment more practical. We measured each patient's vulnerability to membership inference attack, across 7,789 models trained on six clinical datasets in four modalities, including two adapted pretrained medical models. Training choices