“This talk provides a step-by-step introduction on how to use deep learning to perform AES side-channel attacks. After providing a brief overview of what side channels and deep-learning are, we walk you through how to use Tensorflow to build an end-to-end attack that will recover TinyAES keys from SMT32F415 chips using deep learning. Along the way we will discuss what work and what doesn’t based on our experience attacking many hardware AES implementations over the last few years.”
This talk was given at Defcon 27 together with Elie Bursztein.