Using Artificial Intelligence to detect threats without Decryption

No ratings

Presented at FutureConDallas 2019 by

Drew Bradshaw

Networks are becoming more and more opaque as network based encryption is on the rise protecting our traffic as it travels over trusted and untrusted networks. We are not the only ones making use of this as adversaries are hiding their communication in the same encrypted channels. In 2016, Cisco researchers published a paper and an open-source implantation debuting Encrypted Traffic Analytics that is an application of Machine Learning to detect malicious activity without the need for decryption. In this session we will discuss Encrypted Traffic Analytics’ origin, the Machine Learning techniques, and the results we have experienced as it has been implemented in solutions across public and private sectors.