Dark Reading is part of the Informa Tech Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them.Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

Security Management

7/23/2019
07:30 AM
Larry Loeb
Larry Loeb
Larry Loeb
50%
50%

Skylight Says Cylance Can Be Bypassed; Cylance Says, 'Not So Fast, There!'

Skylight's findings certainly gained attention, but at what cost to the company's credibility?

An Australian security firm named Skylight Cyber released a report on July 18 called "Cylance, I kill you!" that asserted that Blackberry's Cylance AV product could be simply bypassed. Before the issuing of the report, they had not informed Cylance of their findings as most researchers that follow "responsible disclosure" protocol will do. Such a protocol is usually followed to avoid the appearance of a conflict of interest, such as an effort to short the affected company's stock.

Skylight did actually find some things out, however. Cylance has long purported that its AI-based approach to endpoint security is superior to past approaches. Cylance says, "artificial intelligence (AI) can truly predict attacks by weeks, months, or even years ahead of a campaign. We call this the Cylance Predictive Advantage (CPA) that AI gives the defender over the attacker." This is just the point that Skylight found a way to bypass.

Skylight found that by taking strings from a non-malicious file and appending them to a malicious one, it could trick the CPA into thinking the malicious file is benign.

The benign strings they used came from an online gaming program (which Skylight has not yet specified). Evidently the Cylance product had been trained during development to ignore this particular program, perhaps due to too many false positive results. No changes to the actual AV code were made.

Skylight tested the method on WannaCry ransomware, Samsam ransomware, Mimikatz, as well as other known malicious files. Most of them (84%) were misidentified during function.

The PROTECT product is not alone in using AI for endpoint security. Symantec, Crowdstrike and Darktrace are also AI proponents. Skylight did not test its method on these products before releasing the report. The method Skylight found may not be effective with these, since they do not use AI to the same depth that PROTECT does.

Skylight thinks that the Cylance product's use of a whitelist is not directly a deadly flaw in it, but not using backup signatures or heuristics to doublecheck the algorithm's conclusion made it so.

Cylance came up with a response over the weekend. It admits the problem, saying "This vulnerability allows the manipulation of a specific type of feature analyzed by the algorithm that in limited circumstances will cause the model to reach an incorrect conclusion."

So, what are they going to do in response? They say that, "First, we have added anti-tampering controls to the parser in order to detect feature manipulation and prevent them from impacting the model score. Second, we have strengthened the model itself to detect when certain features become proportionally overweight. Lastly, we have removed the features in the model that were most susceptible to tampering." In short, Cylane says that it has fixed the puppy. Not only that, it also says that it is able to automatically deploy these enhancements from use of its cloud architecture.

So, Skylight did have a point in its post. It also went around Cylance in a major way, and was completely irresponsible in it they disclosed its findings. Skylight may have gotten the attention it sought, but at the expense of its credibility.

— Larry Loeb has written for many of the last century's major "dead tree" computer magazines, having been, among other things, a consulting editor for BYTE magazine and senior editor for the launch of WebWeek.

Comment  | 
Print  | 
More Insights
Comments
Threaded  |  Newest First  |  Oldest First
COVID-19: Latest Security News & Commentary
Dark Reading Staff 9/25/2020
Hacking Yourself: Marie Moe and Pacemaker Security
Gary McGraw Ph.D., Co-founder Berryville Institute of Machine Learning,  9/21/2020
Startup Aims to Map and Track All the IT and Security Things
Kelly Jackson Higgins, Executive Editor at Dark Reading,  9/22/2020
Register for Dark Reading Newsletters
White Papers
Video
Cartoon
Current Issue
Special Report: Computing's New Normal
This special report examines how IT security organizations have adapted to the "new normal" of computing and what the long-term effects will be. Read it and get a unique set of perspectives on issues ranging from new threats & vulnerabilities as a result of remote working to how enterprise security strategy will be affected long term.
Flash Poll
How IT Security Organizations are Attacking the Cybersecurity Problem
How IT Security Organizations are Attacking the Cybersecurity Problem
The COVID-19 pandemic turned the world -- and enterprise computing -- on end. Here's a look at how cybersecurity teams are retrenching their defense strategies, rebuilding their teams, and selecting new technologies to stop the oncoming rise of online attacks.
Twitter Feed
Dark Reading - Bug Report
Bug Report
Enterprise Vulnerabilities
From DHS/US-CERT's National Vulnerability Database
CVE-2020-15208
PUBLISHED: 2020-09-25
In tensorflow-lite before versions 1.15.4, 2.0.3, 2.1.2, 2.2.1 and 2.3.1, when determining the common dimension size of two tensors, TFLite uses a `DCHECK` which is no-op outside of debug compilation modes. Since the function always returns the dimension of the first tensor, malicious attackers can ...
CVE-2020-15209
PUBLISHED: 2020-09-25
In tensorflow-lite before versions 1.15.4, 2.0.3, 2.1.2, 2.2.1 and 2.3.1, a crafted TFLite model can force a node to have as input a tensor backed by a `nullptr` buffer. This can be achieved by changing a buffer index in the flatbuffer serialization to convert a read-only tensor to a read-write one....
CVE-2020-15210
PUBLISHED: 2020-09-25
In tensorflow-lite before versions 1.15.4, 2.0.3, 2.1.2, 2.2.1 and 2.3.1, if a TFLite saved model uses the same tensor as both input and output of an operator, then, depending on the operator, we can observe a segmentation fault or just memory corruption. We have patched the issue in d58c96946b and ...
CVE-2020-15211
PUBLISHED: 2020-09-25
In TensorFlow Lite before versions 1.15.4, 2.0.3, 2.1.2, 2.2.1 and 2.3.1, saved models in the flatbuffer format use a double indexing scheme: a model has a set of subgraphs, each subgraph has a set of operators and each operator has a set of input/output tensors. The flatbuffer format uses indices f...
CVE-2020-15212
PUBLISHED: 2020-09-25
In TensorFlow Lite before versions 2.2.1 and 2.3.1, models using segment sum can trigger writes outside of bounds of heap allocated buffers by inserting negative elements in the segment ids tensor. Users having access to `segment_ids_data` can alter `output_index` and then write to outside of `outpu...