2 years of lessons part 2

In Part 1 of this blog post, we discussed the increasing number of smart device data breaches that we’ve uncovered over the past two years, highlighting the need for tighter security. We also made a bold statement, asserting that our approach is superior to anything currently available in the marketplace.

How can we be so sure? First, because we know that standard solutions engage in a dubious cat and mouse game where the good guys build walls and moats, only to have the bad guys find new ways to hack in. New “upgrades” are issued, and the cycle repeats itself.

Our solution is different, it’s better, and it never becomes obsolete. Simply put, we sample smart device data going from your home to the cloud so that we can understand what your devices are doing.

Technically speaking, our data science team built sophisticated AI algorithms and advanced data engineering to analyze anonymized, crowdsourced device behavior data in real time. We only look at meta data (not the contents) to analyze transmission patterns and determine exactly how a device should behave.

Even though we’re still in Beta, we’re in lots of homes. We’ve been able to look across the Everything Set community of homes to build a treasure trove of information and analysis about device behaviors -- over a million hours of devices communicating back and forth to the cloud. We’ve developed a continuously updating library of thousands of device behavior profiles, making it possible for us to identify the precise moment a particular smart device starts to display irregular or even dangerous communication patterns.

When smart devices behave in a surprising way, it could mean they have been infected by an unwanted and unexpected source (whether person or program) and are sharing information from the home to an unintended or dangerous electronic location. At that point, your privacy and/or data security might be compromised. A simple example would be someone receiving private transmissions from your home video cameras or being sent your personal data without your consent.

Our AI and data science can immediately recognize deviant data transmission patterns. We can then stop, block, or help the user fix or replace the misbehaving device. Speed is of the essence. The longer an intruder has access to your smart devices, the more damage they can do. By quickly alerting our users, we can prevent catastrophe.