Described as being equivalent to the invention of the light bulb, a standalone battery-free, solar-powered AI technology has been developed by Xnor.
Typically, devices such as smartphones that utilize AI aren’t actually processing it themselves but rather use a cloud due to the large computational power required. With the cloud dependancy and immense energy required to run the algorithms limiting widespread AI use in every life, Xnor’s standalone AI technology enables processing with the central processing unit (CPU) without the need of a cloud.
“The carbon footprint of data centers running all of those algorithms is a key issue,” Farhadi said. “And with the way AI is progressing, it will be a disastrous issue pretty soon, if we don’t think about how we’re going to power our AI algorithms. Data centers, cloud-based solutions for edge-use cases are not actually efficient ways, but other than efficiency, it’s harming our planet in a dangerous way.”
According to a paper by the Natural Resources Defense Council, U.S. data centers consumed an estimated 91 billion kilowatt-hours of electricity in 2013. Projected to increase to roughly 140 billion kilowatt-hours annually by 2020, it will cost $13 billion annually in electricity bills and emit 100 million metric tons of carbon pollution per year.
“Power will become the biggest bottleneck to scaling AI” added Farhadi. “What Xnor has proved today is that it is now possible to run AI inference at such low power that you don’t even need a battery. This will change not only the way products are built in the future, but how entire cities and countries deploy AI solutions at scale.”
As a proof of concept, an AI hardware device fitted with a 320×240 resolution camera capable of object recognition in real time has been created.
“Our homes can be way smarter than they are today. Why? Because now we can have many of these devices deployed in our houses,” explained Farhadi. “It doesn’t need to be a camera. We picked a camera because we wanted to show that the most expensive algorithms can run on this device. It might be audio. … It might be a way smarter smoke detector.”
With the AI hardware device made up of components that cost roughly $10 in total, Xnor’s head of hardware engineering, Saman Naderiparizi, says that the device can operate continuesly for 32 years on solar power alone. It is also fitted with a supercapacitor so that it can store energy and operate even when the solar panel is obscured.
“This is a key technology milestone, not a product,” said Farhadi.
To learn more, click the link below.
