In BriefIntel is releasing a new microprocessor before the year ends, and it's designed to run artificial neural networks better than today's computer chips. The Nervana, as Intel calls it, is going to revolutionize AI computing, especially for businesses.
There’s no denying it, artificial intelligence (AI) is taking over how companies do business, with machine learning powering more and more systems and products — from mobile phones to online services. American tech giant Intel wants to be ahead of this shift, and so they’ve developed a number of technologies designed to advance AI capabilities — including a microprocessor specially designed to run artificial neural networks.
Formerly codenamed Lake Crest, Intel’s first-generation silicon processor for neural networks is called the Nervana Neural Network Processor (NNP), which is powerful enough to handle the intensive computational requirements of running deep neural nets.
“We have multiple generations of Intel Nervana NNP products in the pipeline that will deliver higher performance and enable new levels of scalability for AI models,” CEO Brian Krzanich wrote in a press announcement. Krzanich first announced the Nervana during The Wall Street Journal’s D.Live event on Tuesday. He continued in the announcement that “This puts us on track to exceed the goal we set last year of achieving 100 times greater AI performance by 2020.”
Intel plans to ship this hardware to a small number of its partners before the year ends, but it’s also easily accessible through the company’s Nervana Cloud Service. With this processor, Intel “promises to revolutionize AI computing across myriad industries,” Krzanich added. The company plans to apply the technology to health care, the automotive sector, weather services, and social media. Alongside the Nervana, Intel has also made advances in neuromorphic and quantum computing.