Navigating the Data Labeling Bottleneck as Security Embraces AI

Security now prefers statistics over signatures to fend off attacks. But red teams haven’t yet transitioned from man vs. machine to machine vs. machine, despite a data labeling bottleneck that makes waging AI-based attacks easier than building AI-based defenses. Before attackers catch on, presenters will make the case for AI-based white hat hacking, which will help preemptively plug up the industry’s holes.

Learning Objectives:
1: Learn why the data labeling bottleneck is a big problem for security.
2: Hear about detailed AI-based attacks, less prone to the data labeling bottleneck.
3: Learn methods that’ll help close the gap between offensive and defensive AI.



Hyrum Anderson Technical Director of Data Science, Endgame


Philip Tully Principal Data Scientist, ZeroFOX

← View more Videos

This document was retrieved from on Tue, 26 Mar 2019 06:56:09 -0400.