Intel Showcases AI Strategy for 5G, Intelligent Edge

LAS VEGAS — Intel’s presentation at CES 2020 set the stage for its strategy around three emerging technologies — 5G, artificial intelligence (AI), and the intelligent edge. And it focused on how it plans to use AI to process, store, and move data around networks with more efficiency and insights. LAS VEGAS — Intel’s presentation at CES 2020 set the stage for its strategy around three emerging technologies — 5G, artificial intelligence (AI), and the intelligent edge. And it focused on how it plans to use AI to process, store, and move data around networks with more efficiency and insights. “These technologies are not standalone technologies, they’re actually reinforcing,” Intel CEO Bob Swan said. “5G is enabling intelligent devices, but intelligent devices are putting challenges on developing the technologies for AI and for 5G, so they’re not independent technology inflections,” he said. “The challenge and our role is to fuel silicon technology and software at the core of each and every one of these technologies.” Intel expects connected devices will create 175 zettabytes of data by 2025, growing at an annual growth rate of 25% for at least the next five years. “Over this time frame over 50% of that data will be created by devices at the edge, spanning smart devices, wearables, smart industrial factories, technology being deployed to make retail shopping experiences greater, and connected cars,” Swan said. While half “of the data that will be created will be from these billions of connected, intelligent devices at the edge,” Swan noted that “only 1% of that data today is being processed by artificial intelligence models.” Intel and its partners are “just getting started in terms of the amount of data that’s being created and leveraged to create new and better experiences around the globe,” he said. “5G increasingly is a part and required because of the billions and billions of connected devices at the edge. And AI is making 5G and this convergence of compute to communications to be more and more relevant,” Swan explained. “These billions of devices at the edge need to be able to have compute closer to them.” All of that data needs to be processed to make it more relevant and useful, stored and made readily accessible in perpetuity, and moved faster within and between networks so that it can be processed again, Swan said. As such, Intel is on a mission to embed AI into everything it makes. “We’re simply embedding intelligence across everything that we do” by investing organically and through acquisitions, he explained, perhaps referencing Intel’s recent $2 billion purchase of AI-focused startup Habana Labs. Intel also teased some new processors and a graphics processing unit (GPU) that are in the early stages of development. Details were light in a few cases and some of the products may not ship until later this year or early 2021, but it’s clear that Intel felt some pressure to respond to AMD’s flurry of new products announced hours earlier. The company’s third-generation Intel Xeon Scalable processors, which are coming in the first half of 2020, will feature built-in AI training acceleration and up to a 60% increase in training performance over the previous generation of processors for data centers, according to Intel. EVP Gregory Bryant also introduced Intel’s next-generation mobile processors, code-named Tiger Lake. The CPUs are built on the 10 nanometer (nm) semiconductor manufacturing process, and they feature support for Intel’s Xe graphics architecture, Thunderbolt 4, and WiFi 6. Laptops equipped with Tiger Lake processors are expected to ship later this year, according to Intel. Intel also previewed a forthcoming Xe-based GPU code-named DG1, but the company didn’t share many details outside of an on-stage gaming demo and it isn’t expected to ship in laptops targeted to gamers until late 2020.