These smart sensors warn of disaster before it happens

They don't look like much, but the software that powers these smart home sensors is pretty intriguing.

Ry Crist Senior Editor / Reviews - Labs
Originally hailing from Troy, Ohio, Ry Crist is a writer, a text-based adventure connoisseur, a lover of terrible movies and an enthusiastic yet mediocre cook. A CNET editor since 2013, Ry's beats include smart home tech, lighting, appliances, broadband and home networking.
Expertise Smart home technology | Wireless connectivity Credentials
  • 10 years product testing experience with the CNET Home team
Ry Crist
2 min read

There are all sorts of smart home sensors that promise to keep an eye on your home. At first glance, a new system called OnePrevent doesn't look much different from any of them (other than the steeper-than-average asking price of $299). In fact, OnePrevent looks pretty basic and uninspired.


Looks can be deceiving, though. OnePrevent is backed by a custom-built algorithm called OneEvent. The algorithm was developed by a small startup based out of Wisconsin that's backed by investments that total $4.3 million. And even though it will cost you a subscription fee of $25 per month, the OneEvent pitch is pretty intriguing -- it claims it can detect household disasters like fires and floods up to 20 minutes before they actually happen.

That might sound like something out of 'Minority Report,' but it actually comes from a software-based approach that uses sensors to monitor things like heat and humidity in key areas of your home. If things start to deviate from the norm because of a leaky pipe or an oven you forgot to turn off, the system will catch it, let you know about it, and, most interestingly, learn from it.

"There's a different signature for burnt toast than there is for a smoldering Christmas tree," said Avi Rosenthal, Internet of Things Adviser for OneEvent. According to Rosenthal, users have the option of telling the system what actually happened after being alerted to changing conditions. The OneEvent algorithm aggregates that user data; the idea is that it'll get smarter and smarter as more people use the OnePrevent sensors.

OneEvent's team even went as far as to partner with local fire officials in Wisconsin who were running controlled burns of condemned properties. By sticking OnePrevent sensors throughout the homes before they were set ablaze, OneEvent said it was able to teach the algorithm a great deal about recognizing when a fire might be about to break out.


OnePrevent's app promises to warn you of in-home disasters before they get truly out of hand.


Of course, preemptive alerts about potential disasters could mean a fair share of false positives, and Rosenthal said the system can take a few weeks to learn what normal conditions are in your home.

"Once the system settles in, it is very good," he said.

Available now through OneEvent's website and a network of dealers, the OnePrevent system includes smoke and carbon monoxide detectors, along with sensors for heat and humidity, motion sensors and open/closed detectors for doors and windows. At its base price of $299, the kit is much more expensive than similar kits from smart-home mainstays like Insteon and SmartThings, and the monthly subscription fees certainly don't help, either.

Still, it's an interesting glimpse at how machine learning and AI could soon begin to take better advantage of the ever-growing ocean of data generated by today's smart-home gadgets. I'm especially curious to see if OneEvent ultimately catches the eye of any larger third parties who might want to put that software to use in their own systems. Time will tell -- but for now, watch this space.