IoT

IoT Physical Attack Exploit to be Revealed at Black Hat

Security researcher Billy Rios plans to demonstrate how an exploit can cause an IoT device to launch a physical attack against a human.

IoT devices are increasingly becoming ubiquitous, raising the stakes of physical harm to humans if exploits make these connected devices go rogue. One of the first examples of such an IoT exploit that will do just that is slated to be presented by renowned researcher Billy Rios later this month at Black Hat USA in Las Vegas.

Rios, founder of WhiteScope, in his Black Hat talk When IoT Attacks: Understanding the Safety Risks Associated with Connected Devices will demontrate an exploit he wrote that prompts an IoT device to intentionally strike a person. "We are in the early stages of IoT, but in 15 years robotics will be in our homes and workplace and we want to show the road we are headed on," Rios says.

His exploit employs zero-day vulnerabilities that he discovered in an IoT device, the details of which Rios declined to divulge prior to his presentation. The vendor of the IoT product has not yet patched the flaws, which involve authentication bypass and the ability to disable or bypass safety mechanisms.

 

An IoT physical attack could involve a connected device that strikes an individual in a public place, for example, he says.

"A robotic arm in a factory can hit you, but people don't take that seriously because they think that they are only used in manufacturing," Rios explains. "The attacks I will be discussing are devices that are used in public places and can hit or strike you."

While the number of exploitable connected devices is high, the number of connected devices that can be exploited to physically injure someone is surprisingly small. "I think drones and self-driving cars fit into this category," Rios says.

In the future, however, he anticipates more IoT physical risks, especially as a result of the growth in robotics.

"We have already shown we can hurt people by accident with IoT," Rios says, pointing to recent accidents in self-driving cars. 

Core IoT Problems

Although IoT devices have the potential to physically inflict harm on humans if they are exploited by nefarious actors, the industry is currently unregulated, compared to the transportation and healthcare industries, Rios says.

He added one of the goals of his presentation is to get people talking about a cybersecurity safety law and also the need to activate a scoring system for IoT safety risks, similar to that issued for security update patches.

"A vulnerability that compromises a TV and a vulnerability that compromises a car are currently scored the same way," Rios says. "But where this falls down is when you have a device that can actually hurt you. You want to differentiate between an issue that can physically hurt you and one that doesn't."

Related Content:

Dawn Kawamoto is an Associate Editor for Dark Reading, where she covers cybersecurity news and trends. She is an award-winning journalist who has written and edited technology, management, leadership, career, finance, and innovation stories for such publications as CNET's ... View Full Bio

Comment  | 
Print  | 
More Insights
Comments
Newest First  |  Oldest First  |  Threaded View
Joe Stanganelli
50%
50%
Joe Stanganelli,
User Rank: Ninja
7/8/2017 | 8:40:58 AM
Hurtables
Related: Attacks against connected pacemakers (as seen in Homeland; indeed, it was for this reason that VP Dick Cheney's pacemaker connectivity was disabled).

Nearly four years ago, cybersecurity firm predicted the first IoT-enabled murder to be committed by the end of 2013. I don't think they were correct, but the day is perhaps coming.

I wrote a piece about "murderables" and other cybercrime-as-a-service possibilities at the time, here: newipagency.com/author.asp?section_id=325&doc_id=711779
Register for Dark Reading Newsletters
White Papers
Video
Cartoon Contest
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-2018-6345
PUBLISHED: 2019-01-15
The function number_format is vulnerable to a heap overflow issue when its second argument ($dec_points) is excessively large. The internal implementation of the function will cause a string to be created with an invalid length, which can then interact poorly with other functions. This affects all s...
CVE-2018-7603
PUBLISHED: 2019-01-15
In Drupal's 3rd party module search auto complete prior to versions 7.x-4.8 there is a Cross Site Scripting vulnerability. This Search Autocomplete module enables you to autocomplete textfield using data from your website (nodes, comments, etc.). The module doesn't sufficiently filter user-entered t...
CVE-2019-3554
PUBLISHED: 2019-01-15
Wangle's AcceptRoutingHandler incorrectly casts a socket when accepting a TLS 1.3 connection, leading to a potential denial of service attack against systems accepting such connections. This affects versions of Wangle prior to v2019.01.14.00
CVE-2019-3557
PUBLISHED: 2019-01-15
The implementations of streams for bz2 and php://output improperly implemented their readImpl functions, returning -1 consistently. This behavior caused some stream functions, such as stream_get_line, to trigger an out-of-bounds read when operating on such malformed streams. The implementations were...
CVE-2019-0030
PUBLISHED: 2019-01-15
Juniper ATP uses DES and a hardcoded salt for password hashing, allowing for trivial de-hashing of the password file contents. This issue affects Juniper ATP 5.0 versions prior to 5.0.3.