Exploiting ML models with pickle file attacks: Part 1

By Boyan Milanov We’ve developed a new hybrid machine learning (ML) model exploitation technique called Sleepy Pickle that takes advantage of the pervasive and notoriously insecure Pickle file format used to package and distribute ML models. Sleepy pickle goes beyond previous exploit techniques that target an organization’s systems when they deploy ML models to instead […]

attacks beyond called deploy exploit exploitation exploiting file goes hybrid insecure machine machine learning ml models organization package pickle systems target techniques