Heterogeneous computing and the future of AI – News

Heterogeneous computing is quietly redefining how AI shows up in everyday devices, shifting intelligence closer to where people actually use it rather than keeping it locked inside distant cloud systems.

As AI moves into physical products, success depends less on raw performance and more on how CPUs, GPUs and NPUs work together efficiently. Heterogeneous computing shows up as faster-feeling AI, better battery life and intelligence that responds in context without constantly calling attention to itself, according to Chris Bergey (pictured), executive vice president of the Edge AI Business Unit at Arm.

“Our partnership with Nvidia has been super strong, and it really comes from, in this AI world, the blending of computing and acceleration — and you can see that across any of these devices,” Bergey said. “In these cell phones that have an Arm CPU, an Arm GPU and also an NPU, it’s really that heterogeneous computing and how are we using that to further acceleration of what’s possible on AI.”

Bergey spoke with Savannah Peterson at CES 2026, during an exclusive broadcast on theCUBE, News Media’s livestreaming studio. They discussed how heterogeneous computing enables AI to run efficiently at the edge, transforming devices, wearables, security and developer experiences.

How heterogeneous computing is changing AI at the edge

Heterogeneous computing is becoming essential as AI workloads grow more complex and interactive. Rather than forcing every task through a single processor, workloads are distributed across specialized components — each doing what they do best. That approach enables real-time responsiveness while keeping power consumption in check, Bergey explained.

“When you have those three working together, you can optimize your battery life in a different way,” he said. “You can really optimize every feature in action that happens on that phone rather than just draining from the same thing all at once.”

This architectural shift also changes how developers think about building AI-driven applications. AI is no longer something bolted onto a product after the fact; it becomes a core design consideration. As intelligence moves closer to the user, software must be written to take advantage of multiple compute paths working in concert.

“One of the biggest challenges is actually programming and programming AI,” Bergey said. “How do you program GPUs? How do you program NPUs? That’s something we’re very focused on.”

Beyond smartphones and consumer electronics, heterogeneous computing is opening the door to entirely new classes of devices. Wearables, in particular, are emerging as a proving ground for what happens when AI processing lives directly on the body. These systems collect richer data while maintaining privacy and responsiveness by processing information locally, Bergey noted.

“AI is enabling kind of a renaissance of wearables,” he said. “We’re starting to interact with these devices now. It gives you capabilities in these wearables that wasn’t possible.”

The long-term impact extends beyond convenience into areas such as health, safety and quality of life. As devices gain the ability to process biometric and contextual data in real time, early detection and personalized insights become realistic outcomes rather than distant promises. The key challenge is ensuring that this intelligence is paired with trust and security at the architectural level.

“This is super powerful, but it needs to be something that also meets all the security requirements,” Bergey added. “We’re taking security just at its distinction and all kinds of things you can do to really be able to give that capability at the building block level of really complete trust.”

Here’s the complete video interview, part of News’s and theCUBE’s coverage of CES 2026:

Photo: News