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.

Attacks/Breaches

3/25/2016
11:30 AM
Andrew Thomson
Andrew Thomson
Commentary
Connect Directly
Twitter
LinkedIn
RSS
E-Mail vvv
50%
50%

How 4 Startups Are Harnessing AI In The Invisible Cyberwar

Cybersecurity startups are setting their scopes on a potential goldmine of automated systems they hope will be more effective than hiring human enterprise security teams.

There is growing concern across the board that we might be losing control over cybersecurity. The rapid changes in how we use technology to communicate and the increased number of connected devices means the points of entry or breach are growing. Because the pace of change has been so rapid, security hasn't adapted fast enough and hackers are taking full advantage. The traditional ways of dealing with cyber threats are beginning to look hopelessly inadequate.

This concern goes right to the top. Since entering the White House in 2009, President Obama has repeatedly called for improvements in cybersecurity and in December 2015 announced a new cybersecurity bill which allocated $14 billion of federal spending to further secure government information online. With global cyber spending expected to reach $170 billion by 2020, eyes are on the cybersecurity industry to see who can offer the best solutions.

But while the rest of the industry gets up to speed, a number of forward-thinking cybersecurity startups are attempting to harness the power of artificial intelligence to strengthen the defenses of the good guys. They are identifying, locating, and destroying potential threats in a manner that promises to be quicker and more effective than traditional methods.

Darktrace

Using machine learning techniques inspired by the self-learning intelligence of the human immune system, UK-based startup Darktrace tackles the challenge of detecting previously unidentifiable cyber threats in real time, and allows them to be eradicated more quickly than traditional approaches. Unlike traditional cybersecurity systems in which malicious threats and viruses are manually added to a list and then blocked, Darktrace uses a system based on machine learning and mathematics that can detect threats without any prior knowledge of what it is looking for, cutting out the need for human intervention. The groundbreaking new system was developed by engineers and mathematicians from the University of Cambridge.

Jask

JASK, a San Francisco-based startup, is building what it calls “the world’s first, predictive security operations center" for enterprise-level cybersecurity. The system aims to assist enterprises of all sizes keep ahead of sophisticated cyberattackers by moving past the limitations of existing solutions with proactive A.I security measures. With enterprises adding more and more software applications to their networks, and relying more heavily on cloud for saving data, JASK’s approach  “finds threats buried in data — all in an automated way that doesn’t require [companies] to throw more bodies at the problem,” according to Greg Martin, Jask founder and CEO.

Deep Instinct

Launched in November 2015, this Tel Aviv-based startup is using sophisticated deep learning algorithms to improve cybersecurity in the banking, financial, and government spheres in the U.S and Israel. The Deep Instinct engine is modeled on the human brain’s ability to learn. Once a brain learns to identify an object, it can identify it again in the future instinctively. Similarly, as Deep Instinct’s artificial brain learns to detect any type of cyber threat, its prediction capabilities become faster and more developed. The company recently partnered with FireLayers to create the first commercially available AI solution for enterprise cloud applications. The solution focuses on both detection and prevention, targeting the market for advanced persistent threats (APT) solutions.

harvest.ai

harvest.ai is approaching AI cybersecurity from a slightly different angle, based on the idea that to truly secure your defenses, you need to know your weak points and principle targets.

Founder and CEO Alexander Watson is no stranger to industrial espionage and cyberattacks, having worked as a field agent for the NSA for nearly a decade. The company has created AI-based algorithms that learn the business value of critical documents, monitor who is using or moving them, and detect and stop data breaches from targeted attacks and insider threats before data can be copied or stolen. harvest.ai’s MACIE system detects anomalies in how users access the network by analyzing changes in location of access, browsing habits, data transfers and other telemetry that can be harnessed from external systems. The system can also alert users if an important document is accidentally shared publicly on a cloud or network, or sent to the wrong person. 

Related content:

 

Interop 2016 Las VegasFind out more about security threats at Interop 2016, May 2-6, at the Mandalay Bay Convention Center, Las Vegas. Click here for pricing and to register. 

Andrew Thomson is the CEO and Founder of VentureRadar, a big data and machine-learning company that discovers and ranks companies, making them visible to potential partners, customers, and investors. In recent months the company has been commissioned to do various scouting ... View Full Bio
 

Recommended Reading:

Comment  | 
Print  | 
More Insights
Comments
Oldest First  |  Newest First  |  Threaded View
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...