Threat Intelligence

1/17/2018
10:30 AM
Justin Fier
Justin Fier
Commentary
Connect Directly
Twitter
LinkedIn
RSS
E-Mail vvv
0%
100%

How AI Would Have Caught the Forever 21 Breach

Companies must realize that the days of the desktop/server model are over and focus on "nontraditional" devices.

After discovering that multiple point-of-sale (POS) devices were breached nationwide, retailer Forever 21 joined the list of big-name corporations that suffered a cyberattack in 2017. And because the investigation is still ongoing, it is likely that we won't know the full impact of the incident — including how many people are affected — for months.

However, as the initial details of the breach emerge, the headlines tell a familiar story. Many of the breaches of the past few years share a common theme: abnormal activity had occurred on the network, missed by the organization and having bypassed all of its security tools. How can we proactively identify and tackle these threats as we move into 2018?

As a first step, we must recognize that the days of the desktop/server model are over. In the case of Forever 21, the POS devices served as ground zero — not a laptop, a server, or even a corporate printer. In the age of the Internet of Things, we increasingly rely on "nontraditional" devices to optimize efficiency and boost productivity. But what constitutes a nontraditional device, and how do we look for it? Is it a device without a monitor? A device without a keyboard?

Today a nontraditional device could be anything from heating and cooling systems to Internet-connected coffee machines to a rogue Raspberry Pi hidden underneath the floorboards. Protecting registered corporate devices is not enough — criminals will look for the weakest link. As our businesses grow in digital complexity, we have to monitor the entire infrastructure, including the physical network, virtual and cloud environments, and nontraditional IT, to ensure we can spot irregularities as they emerge.

A subtle irregularity in device behavior is almost always the first sign of an emerging cyber attack — but these early indicators are consistently missed by tools that are rigidly programmed to spot known vulnerabilities and malicious behaviors.

With Forever 21, the encryption technology on the POS devices had failed, but only on some devices. Artificial intelligence (AI) would spot this type of anomaly, even if it had never seen it before, because it learns what normal behavior is over time, using this understanding to recognize suspicious shifts in activity when they arise. In contrast, tools that scan known devices, looking for known viruses or published indicators of compromise, would have missed it.

No matter how large our team is, as security professionals we all face the challenge of finding the evasive needle in an ever-expanding haystack. AI's promise is to make subtle connections and correlations behind the scenes, and constantly build up an understanding of our digital environments over time — with this knowledge getting better and better.

Furthermore, an AI system today can be up and running in minutes, meaning that it can very quickly deliver results. This doesn't just mean catching new anomalous activity but also understanding if a threatening presence is already in operation in your network. How is a cluster of POS devices behaving in comparison with what the AI has learned to be normal for similar devices?

Shifting our teams away from alert-chasing and perimeter protection and toward a workflow focusing on the anomalies found by AI might help us bring a gun to the knife fight. Had Forever 21 been equipped with such technology, it would have had a very good chance of both identifying and remediating the situation before any of its data was compromised.

Indeed, the gap between the breach happening and its disclosure points to a woeful inadequacy in our ability to see and detect emerging problems. Transferring the analytic burden to machines will give human security teams the time to improve their skills and add new ones — focusing on investigating and remediating genuine threats, while also having time to dedicate to strategic initiatives. As things stand, security teams are often caught in a vicious circle: high level-changes need to be made to prevent low-level problems, but teams are so busy fighting fires that they don't have the time to make the changes necessary to break this cycle. AI would give both large and small security teams the ability to break out of this cycle.

Protecting against the threats we know of in advance is no longer sufficient. AI offers the best chance to catch breaches like the one that affected Forever 21, because it looks at all activity, irrespective of whether it pertains to a cash register or a data server, and isn't biased to find threats that it knows already. AI is forever learning — something Forever 21 should bear in mind as it revises its security strategy. 

Related Content:

Justin Fier is the Director for Cyber Intelligence & Analytics at Darktrace, based in Washington, DC. With over 10 years of experience in cyber defense, Fier has supported various elements in the US intelligence community, holding mission-critical security roles with Lockheed ... View Full Bio
Comment  | 
Print  | 
More Insights
Comments
Newest First  |  Oldest First  |  Threaded View
How Cybercriminals Clean Their Dirty Money
Alexon Bell, Global Head of AML & Compliance, Quantexa,  1/22/2019
Facebook Shuts Hundreds of Russia-Linked Pages, Accounts for Disinformation
Sara Peters, Senior Editor at Dark Reading,  1/17/2019
Register for Dark Reading Newsletters
White Papers
Video
Cartoon Contest
Write a Caption, Win a Starbucks Card! Click Here
Latest Comment: "He's not that smart. He's running iOS 11 on a 5c."
Current Issue
The Year in Security 2018
This Dark Reading Tech Digest explores the biggest news stories of 2018 that shaped the cybersecurity landscape.
Flash Poll
How Enterprises Are Attacking the Cybersecurity Problem
How Enterprises Are Attacking the Cybersecurity Problem
Data breach fears and the need to comply with regulations such as GDPR are two major drivers increased spending on security products and technologies. But other factors are contributing to the trend as well. Find out more about how enterprises are attacking the cybersecurity problem by reading our report today.
Twitter Feed
Dark Reading - Bug Report
Bug Report
Enterprise Vulnerabilities
From DHS/US-CERT's National Vulnerability Database
CVE-2018-20742
PUBLISHED: 2019-01-24
An issue was discovered in UC Berkeley RISE Opaque before 2018-12-01. There is no boundary check on ocall_malloc. The return value could be a pointer to enclave memory. It could cause an arbitrary enclave memory write.
CVE-2019-6486
PUBLISHED: 2019-01-24
Go before 1.10.8 and 1.11.x before 1.11.5 mishandles P-521 and P-384 elliptic curves, which allows attackers to cause a denial of service (CPU consumption) or possibly conduct ECDH private key recovery attacks.
CVE-2018-17693
PUBLISHED: 2019-01-24
This vulnerability allows remote attackers to execute arbitrary code on vulnerable installations of Foxit PhantomPDF 9.2.0.9297. User interaction is required to exploit this vulnerability in that the target must visit a malicious page or open a malicious file. The specific flaw exists within the con...
CVE-2018-17694
PUBLISHED: 2019-01-24
This vulnerability allows remote attackers to execute arbitrary code on vulnerable installations of Foxit PhantomPDF 9.2.0.9297. User interaction is required to exploit this vulnerability in that the target must visit a malicious page or open a malicious file. The specific flaw exists within the han...
CVE-2018-17695
PUBLISHED: 2019-01-24
This vulnerability allows remote attackers to execute arbitrary code on vulnerable installations of Foxit PhantomPDF 9.2.0.9297. User interaction is required to exploit this vulnerability in that the target must visit a malicious page or open a malicious file. The specific flaw exists within the han...