Skip to Content
Artificial intelligence

Uptake is putting the world’s data to work

By applying AI to industrial data, the startup is minimizing machines’ downtime.
September 12, 2018
Photo of Ganesh Bell speaking at EmTech 2018
Photo of Ganesh Bell speaking at EmTech 2018Jake Belcher

The majority of the world’s data is lazy. Once produced, it lies abandoned in the abyss of the world’s databases: according to consultancy IDC, less than 0.5 percent of the world’s data is ever used.

“Machines are getting connected at an increasingly higher rate,” Ganesh Bell, the president of Uptake, told the audience at MIT Technology Review’s EmTech conference today. “And those that are already connected are creating a lot of data.”

Most of the time, though, no one’s listening. As Bell put it, it’s as if our machines are posting hundreds of Facebook status updates and tweets a day, but they aren’t getting any likes or comments. Bell wants to increase the interaction that information attracts and put the data to use for a greater purpose.

Bell and his company are pulling out all the untapped data produced in industries like transportation and energy and applying AI to it, hoping to use it to make businesses more efficient.

For example, by looking at energy data, Uptake found that there are more 100 ways for a wind turbine to fail. By training and building a series of algorithms for predicting everything from yaw misalignment to generator failure, they have found they can anticipate more than 90 of those ways. And if you give humans a warning that something is going to go wrong, downtime can be reduced and more energy produced. “We have a mission to build a world that always works,” he said.

For most of Uptake’s clients, there is no need for additional sensors to be added to the machines. It’s just putting the existing data to work.

But as Bell says, this means nothing if humans don’t respond to the warnings. The software only detects that a problem is imminent. It doesn’t fix it. So the company also records how often people respond to the warnings and how accurate they prove to be. This feedback not only ensures that companies are getting the most out of the software; it also provides feedback to Uptake so that it can better train its systems, and continue to improve the industries that keep our cities powered and move us around the world.

Deep Dive

Artificial intelligence

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

Google DeepMind’s new generative model makes Super Mario–like games from scratch

Genie learns how to control games by watching hours and hours of video. It could help train next-gen robots too.

What’s next for generative video

OpenAI's Sora has raised the bar for AI moviemaking. Here are four things to bear in mind as we wrap our heads around what's coming.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.