Vulnerabilities / Threats

7/25/2017
10:00 AM
Connect Directly
Twitter
Twitter
RSS
E-Mail
100%
0%

Using AI to Break Detection Models

Pitting machine learning bots against one another is the new spy vs. spy battle in cybersecurity today.

In the spy versus spy world that pits cybersecurity defense against money-motivated attackers, it always pays to think a few steps ahead of the attackers. Any security pro who has seen the direction that detection is going these days with artificial intelligence and machine learning understands that those next attacks steps are probably going to involve some sort of subversion of the AI algorithms. If security wants to effectively use AI, it will need to find ways to harden those models

This week at Black Hat, one researcher hopes to contribute to the discipline by showing off a new automated AI agent that probes the data science behind machine learning malware detection models and looks for mathematical weaknesses.

"All machine learning models have blind spots. All of them. And a sophisticated and motivated adversary is out there trying to exploit them," says Hyrum Anderson, technical director of data science for Endgame. "We have created an artificial agent that tries to automatically discover those blind spots."

As he puts it, the agent "literally plays a game against our model and tries to beat it," essentially automating the auditing of the mathematical underpinning of detection mechanisms. The agent essentially inspects an executable file and uses a sequence of file mutations to test the detection model. This agent uses its own brand of machine learning to figure out which sequences of mutations are most likely to create a variant that evades the model. Using the information it gains from this automated test, the agent can create a policy for developing malware variants that have a high likelihood of breaking the opposing machine learning model of the detection engine.

This is the logical next step in advancement of thought about cybersecurity's auditing of machine learning efficacy, and one which Anderson hopes to encourage across the industry as vendors further refine their machine learning mechanisms. 

"You'll get no criticisms from me (about the competition). I think in general my colleagues and competitors are all paranoid and are always thinking about how to make (their models) secure. But that's usually a manual process. It's spot-checking and it's somebody looking at it," he says. "We wanted to take that to the next level. I don't believe that our adversaries are yet using this level of sophistication we are proposing in our research, but that's the point. We want to get there before they do."

At the moment, Anderson's use of the agent begins and ends with Endgame's own machine learning model. But, in concert with his presentation, he and his team are going to release code that is generic and adaptable for other vendors and researchers to inspect their own models. 

"We want to put it out there because a rising tide raises all boats," Anderson says.

The idea of machine learning and AI hardening is generally gaining momentum with data scientists and security specialists of late. In fact, this is one of several talks at Black Hat this year that will focus on problems that can arise through flawed machine learning algorithms. For example, in one talk a data scientist with Sophos will discuss how bad data can screw up detection models. Another from a group of Georgia Tech researchers will dive into a new tool that aims to sabotage detection mechanisms in Android antivirus apps. 

 

Related Content:

Ericka Chickowski specializes in coverage of information technology and business innovation. She has focused on information security for the better part of a decade and regularly writes about the security industry as a contributor to Dark Reading.  View Full Bio

Comment  | 
Print  | 
More Insights
Comments
Newest First  |  Oldest First  |  Threaded View
What We Talk About When We Talk About Risk
Jack Jones, Chairman, FAIR Institute,  7/11/2018
Ticketmaster Breach Part of Massive Payment Card Hacking Campaign
Jai Vijayan, Freelance writer,  7/10/2018
7 Ways to Keep DNS Safe
Curtis Franklin Jr., Senior Editor at Dark Reading,  7/10/2018
Register for Dark Reading Newsletters
White Papers
Video
Cartoon Contest
Write a Caption, Win a Starbucks Card! Click Here
Latest Comment: Locked device, Ha! I knew there was another way in.
Current Issue
Flash Poll
Twitter Feed
Dark Reading - Bug Report
Bug Report
Enterprise Vulnerabilities
From DHS/US-CERT's National Vulnerability Database
CVE-2017-15137
PUBLISHED: 2018-07-16
The OpenShift image import whitelist failed to enforce restrictions correctly when running commands such as "oc tag", for example. This could allow a user with access to OpenShift to run images from registries that should not be allowed.
CVE-2017-17541
PUBLISHED: 2018-07-16
A Cross-site Scripting (XSS) vulnerability in Fortinet FortiManager 6.0.0, 5.6.4 and below versions, FortiAnalyzer 6.0.0, 5.6.4 and below versions allows inject Javascript code and HTML tags through the CN value of CA and CRL certificates via the import CA and CRL certificates feature.
CVE-2018-1046
PUBLISHED: 2018-07-16
pdns before version 4.1.2 is vulnerable to a buffer overflow in dnsreplay. In the dnsreplay tool provided with PowerDNS Authoritative, replaying a specially crafted PCAP file can trigger a stack-based buffer overflow, leading to a crash and potentially arbitrary code execution. This buffer overflow ...
CVE-2018-10840
PUBLISHED: 2018-07-16
Linux kernel is vulnerable to a heap-based buffer overflow in the fs/ext4/xattr.c:ext4_xattr_set_entry() function. An attacker could exploit this by operating on a mounted crafted ext4 image.
CVE-2018-10857
PUBLISHED: 2018-07-16
git-annex is vulnerable to a private data exposure and exfiltration attack. It could expose the content of files located outside the git-annex repository, or content from a private web server on localhost or the LAN.