Class Central is learner-supported. When you buy through links on our site, we may earn an affiliate commission.

YouTube

Stop! Sandboxing Exploitable Functions and Modules Using In-Kernel Machine Learning

Black Hat via YouTube

Overview

Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Learn about a new technique for detecting and stopping 0-day exploitations in the Linux kernel through this 28-minute Black Hat presentation. Discover how dynamic sandboxing of exploitable functions and modules can be deployed in scenarios where service interruption and system reboots are unacceptable, while maintaining minimal performance overhead and memory footprint. Explore two key innovations: an eBPF-based runtime checking mechanism ensuring code integrity, data integrity, and argument authentication, and kernel-embedded machine learning models that detect malicious exploitation behaviors. See a demonstration using CVE-2022-0995 as a case study with detailed measurement results. Understand how this technique can be applied when loading device drivers from untrusted vendors, detecting in-the-wild exploits, and preventatively sandboxing low-quality kernel code with vulnerability histories. Presented by researchers from the University of Colorado Boulder, Arizona State University, and Nanjing University.

Syllabus

Stop! Sandboxing Exploitable Functions and Modules Using In-Kernel Machine Learning

Taught by

Black Hat

Reviews

Start your review of Stop! Sandboxing Exploitable Functions and Modules Using In-Kernel Machine Learning

Never Stop Learning.

Get personalized course recommendations, track subjects and courses with reminders, and more.

Someone learning on their laptop while sitting on the floor.