Technology page

IoT security

A chain is only as strong as its weakest link, and that conventional wisdom applies to the connected home as well. Most smart home accessories have lackluster security that leaves their users vulnerable to attack. We're now investing in new cloud-based technology with the aim of being able to protect the expanding IoT world.

In-depth knowledge of the IoT landscape and vulnerabilities

The Internet of Things (IoT) poses significant risk to the entire digital ecosystem. This is because so many of these devices are designed without a built-in security system to keep them from being hijacked by hackers.

As this new space grows, Avast is investing in new technologies with our cloud-based security infrastructure that includes analyzing and processing IoT threat data at the network level.

Our Wi-Fi Inspector technology, deployed since 2015, helps assess and identify vulnerabilities in connected devices on over 50 million home networks. It also helps remedy configuration errors. Rich data combined with AI and machine learning will continue to enrich our ability to detect and protect against IoT threats.

How we protect IoT devices

We study the behavior of IoT devices, by themselves and in a group, to statistically evaluate the amount and types of data they send, and then use this in conjunction with our analysis of the user’s infrastructure. This means we look at what type of devices and network they’re on, and critically study the capabilities and vulnerabilities of each, using our sophisticated Artifical Intelligence engine. With this information and algorithms, we are able to protect your network from any incoming threats.

Avast Smart Home

Statistics from hundreds of millions of devices are processed through the Machine Learning algorithms running in our distributed cloud infrastructure in order to detect threats on connected devices. Our expertise with malware gives us the ability to minimize latency; shortening the critical time needed to identify threats. But to truly combat the increasing array of dangers across an ever-increasing number of device types, we use a two-step approach. First, our custom-built anomaly detection algorithms are used to detect specific attack types: we stitch these together in order to identify different classes of IoT attacks. Second, we engage our deep neural network to identify all these attacks, as well as emerging ones.

A deep neural net for IoT traffic

Our network traffic feeds statistics from millions of homes into a deep neural network. Interior nodes of this complex network then begin to understand how these flows relate to specific device types, how a number of devices make up a home configuration, or how a larger number of devices make up a service. Then, the output or final layer of the neural network separates out collections of flows that are either benign, or malicious.

Our Smart Home architecture

In order to protect our customers and their home IoT devices, Avast is present in both the router and through our existing products for PC, Mac, and mobile devices. These collect statistics about the traffic flows going in and out of the home, without affecting the normal routing of the packets. Collected stats are then sent to the Avast Cloud for processing. If the Avast Intelligence Platform deems the flow or device to be malicious, it will instruct the Avast product to block the affected flow or device. The Avast Smart Home app provides a portal for the user to interact with the system.

Why is Avast positioned for success in IoT?

Simply put, scale matters: with access to data from over 50 million consumer homes in 100 countries, Avast has an unparalleled overview of the threat landscape. This means we can identify and respond to threats faster than anyone else.

Explore other technology fields

Stay informed, stay protected

We continuously monitor security issues to protect our hundreds of millions of users from emerging threats. To get the latest product features first and hear about threats from Avast’s Threat Labs experts, visit the Avast blog.