When Your AI Can't Be Trusted Anymore | IdentityShield '26
What if your ML system is running perfectly—but making the wrong decisions? This talk explores Ransomware 3.0, where attackers poison models and pipelines instead of locking systems, and shows how AI‑augmented attacks bypass traditional security and how to defend against silent ML compromise. Speaker: Avinish Thakur Software Engineer, miniOrange Pune, India.