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.

Application Security

12/27/2018
10:30 AM
Connect Directly
Twitter
LinkedIn
RSS
E-Mail vvv
50%
50%

Toxic Data: How 'Deepfakes' Threaten Cybersecurity

The joining of 'deep learning' and 'fake news' makes it possible to create audio and video of real people saying words they never spoke or things they never did.

"Fake news" is one of the most widely used phrases of our times. Never has there been such focus on the importance of being able to trust and validate the authenticity of shared information. But its lesser-understood counterpart, "deepfake," poses a much more insidious threat to the cybersecurity landscape — far more dangerous than a simple hack or data breach.

Deepfake activity was mostly limited to the artificial intelligence (AI) research community until late 2017, when a Reddit user who went by "Deepfakes" — a portmanteau of "deep learning" and "fake" — started posting digitally altered pornographic videos. This machine learning technique makes it possible to create audio and video of real people saying and doing things they never said or did. But Buzzfeed brought more visibility to Deepfakes and the ability to digitally manipulate content when it created a video that supposedly showed President Barack Obama mocking Donald Trump. In reality, deepfake technology had been used to superimpose President Obama's face onto footage of Jordan Peele, the Hollywood filmmaker.  

This is just one example of a new wave of attacks that are growing quickly. They have the potential to cause significant harm to society overall and to organizations within the private and public sectors because they are hard to detect and equally hard to disprove.

The ability to manipulate content in such unprecedented ways generates a fundamental trust problem for consumers and brands, for decision makers and politicians, and for all media as information providers. The emerging era of AI and deep learning technologies will make the creation of deepfakes easier and more "realistic," to an extent where a new perceived reality is created. As a result, the potential to undermine trust and spread misinformation increases like never before.

To date, the industry has been focused on the unauthorized access of data. But the motivation behind and the anatomy of an attack has changed. Instead of stealing information or holding it ransom, a new breed of hackers now attempts to modify data while leaving it in place.

One study from Sonatype, a provider of DevOps-native tools, predicts that, by 2020, 50% of organizations will have suffered damage caused by fraudulent data and software. Companies today must safeguard the chain of custody for every digital asset in order to detect and deter data tampering.

The True Cost of Data Manipulation
There are many scenarios in which altered data can serve cybercriminals better than stolen information. One is financial gain: A competitor could tamper with financial account databases using a simple attack to multiply all the company's account receivables by a small random number. While a seemingly small variability in the data could go unnoticed by a casual observer, it could completely sabotage earnings reporting, which would ruin the company's relationship with its customers, partners, and investors.

Another motivation is changing perception. Nation-states could intercept news reports that are coming from an event and change those reports before they reach their destination. Intrusions that undercut data integrity have the potential to be a powerful arm of propaganda and misinformation by foreign governments.

Data tampering can also have a very real effect on the lives of individuals, especially within the healthcare and pharmaceutical industries. Attackers could alter information about the medications that patients are prescribed, instructions on how and when to take them, or records detailing allergies.

What do organizations need to consider to ensure that their digital assets remain safe from tampering? First, software developers must focus on building trust into every product, process, and transaction by looking more deeply into the enterprise systems and processes that store and exchange data. In the same way that data is backed up, mirrored, or encrypted, it continually needs to be validated to ensure its authenticity. This is especially critical if that data is being used by AI or machine learning applications to run simulations, to interact with consumers or partners, or for mission-critical decision-making and business operations.

The consequences of deepfake attacks are too large to ignore. It's no longer enough to install and maintain security systems in order to know that digital assets have been hacked and potentially stolen. The recent hacks on Marriott and Quora are the latest on the growing list of companies that have had their consumer data exposed. Now, companies also need to be able to validate the authenticity of their data, processes, and transactions.

If they can't, it's toxic.

Related Content:

Dirk Kanngiesser is the co-founder and CEO of Cryptowerk, a provider of data integrity solutions that make it easy to seal digital assets and prove their authenticity at scale using blockchain technology. With more than 25 years of technology leadership experience, Dirk has ... View Full Bio
Comment  | 
Print  | 
More Insights
Comments
Newest First  |  Oldest First  |  Threaded View
Data Privacy Protections for the Most Vulnerable -- Children
Dimitri Sirota, Founder & CEO of BigID,  10/17/2019
Sodinokibi Ransomware: Where Attackers' Money Goes
Kelly Sheridan, Staff Editor, Dark Reading,  10/15/2019
Register for Dark Reading Newsletters
White Papers
Video
Cartoon
Current Issue
7 Threats & Disruptive Forces Changing the Face of Cybersecurity
This Dark Reading Tech Digest gives an in-depth look at the biggest emerging threats and disruptive forces that are changing the face of cybersecurity today.
Flash Poll
Twitter Feed
Dark Reading - Bug Report
Bug Report
Enterprise Vulnerabilities
From DHS/US-CERT's National Vulnerability Database
CVE-2019-18214
PUBLISHED: 2019-10-19
The Video_Converter app 0.1.0 for Nextcloud allows denial of service (CPU and memory consumption) via multiple concurrent conversions because many FFmpeg processes may be running at once. (The workload is not queued for serial execution.)
CVE-2019-18202
PUBLISHED: 2019-10-19
Information Disclosure is possible on WAGO Series PFC100 and PFC200 devices before FW12 due to improper access control. A remote attacker can check for the existence of paths and file names via crafted HTTP requests.
CVE-2019-18209
PUBLISHED: 2019-10-19
templates/pad.html in Etherpad-Lite 1.7.5 has XSS when the browser does not encode the path of the URL, as demonstrated by Internet Explorer.
CVE-2019-18198
PUBLISHED: 2019-10-18
In the Linux kernel before 5.3.4, a reference count usage error in the fib6_rule_suppress() function in the fib6 suppression feature of net/ipv6/fib6_rules.c, when handling the FIB_LOOKUP_NOREF flag, can be exploited by a local attacker to corrupt memory, aka CID-ca7a03c41753.
CVE-2019-18197
PUBLISHED: 2019-10-18
In xsltCopyText in transform.c in libxslt 1.1.33, a pointer variable isn't reset under certain circumstances. If the relevant memory area happened to be freed and reused in a certain way, a bounds check could fail and memory outside a buffer could be written to, or uninitialized data could be disclo...