SAN DIEGO — Qualcomm executives this week attempted to draw contrasts and explain how their approach to cloud computing, artificial intelligence (AI), and mobile edge computing is different than other technology companies and operators. “Our focus is different than the rest of the industry,” Jim Thompson, CTO and executive vice president at Qualcomm said during a media gathering at the company’s headquarters.

AI, an area of technology that Qualcomm has been working on for the better part of a decade, plays a big role in Qualcomm’s vision for what it calls the “edge cloud,” Thompson explained. “Our focus has been deep neural networks, deep learning for low power, for devices that you would have a battery in and have limited thermal capability,” he said, adding that Qualcomm’s interest is not the cloud essentially.

“AI is very good at consuming large amounts of data. It comes from the edge of the network, it comes from what people do, it comes from sensors, and all of that is at the edge of the network,” he said. “We think that AI is going to have a much bigger impact at the edge of the network, but AI at the edge of the network without connectivity is somewhat limited. So we think that 5G and AI are an incredibly important combination.”

5G can connect a device to the cloud, just like 4G LTE can today, but in 4G LTE networks there’s a big latency gap between a device at the edge and the cloud, he explained. In the 5G era there will be many cases where the cloud edge won’t be necessary but applications or instances that rely on privacy, security, and low latency will make it crucial, he added.

Thompson describes the edge cloud as the cloud and the capability of the cloud, but in a context that’s much closer to the device. “You could do things like offload certain types of computation,” especially in cases where the form factor of the device can’t facilitate the level of computation required because of inherent limitations in heat dissipation,” he explained.

The edge cloud will also act as a go-between for networks in cases where sending massive amounts of data to the cloud is not feasible or even advisable. Qualcomm believes there’s a “partition between the cloud, the edge, and then this very close edge cloud that is milliseconds away from the device,” Thompson said.

“There will be the cloud component — it’s not like it’s going to go away, we’re not saying that,” he said. “We have the edge, remember that’s where all the data is. And then we have this distributed edge cloud that is very close to the device from a latency point of view, so milliseconds from the device versus hundreds of milliseconds from the device.”

That framework is also significantly different from how most people interact with cloud-centric AI today. “Everyone’s experience here to date with AI, it’s really cloud based,” Thompson said. The user clicks around the web, submits images and words, and “then what is happening in the cloud is they’re trying to profile you essentially and predict what it is you will buy and then direct advertising to you. That’s the primary use of AI today in the cloud, but remember all that data didn’t come from you. It came from the edge.”

Photo credit: Qualcomm