IoT and AI are working together to evolve capabilities

AI is everywhere, and the IoT space is no exception—as networks of “smart” devices evolve, they’ll need to incorporate new technologies.

At this year’s CES, we talked to tech experts about how the IoT sector is changing, and how AI fits into it.

Do the evolution. There is a “Cambrian explosion” of AI capabilities right now, Akamai EVP and CTO Bobby Blumofe told IT Brew, and it’s spreading to the IoT space. For professionals in that sector, he said, the potential goes beyond LLMs and chatbots.

“A lot of the real value that we’re getting out of AI, whether it’s to detect fraud, recommend products, drive a car autonomously, or run a robot in a warehouse—in almost all cases, those are not LLMs,” Blumofe said. “Those are deep learning models that are specialized to solving the particular problem at hand.”

Checking in. During a meeting at CES, Silicon Labs CEO Matt Johnson told IT Brew that he is seeing AI integration at every level of the IoT and chip space. The huge amount of data created and consumed by IoT devices is helping developers and manufacturers alike expand their capabilities and continuing to accelerate a tech explosion. And people are getting on board.

“Customers aren’t experts in AI, most of our customers might be experts at washing machines or home security systems or meters, but when it comes to harnessing and mastering their data, a way to actually take advantage of it—it’s a challenge for them,” Johnson said. “But we are seeing it start to happen.”

Ubiquity. That AI has become so normalized in the day to day of the tech industry is a sign of its success, Lenovo Global CIO Arthur Hu told IT Brew. Lenovo’s presentation at CES was heavy on its AI usage and its agreements with Nvidia and AMD, among others. While the company hyped the potential to the conference crowd, Hu said that the real measure of the technology’s success is that it’s fading into the background—like electricity.

“No one ever thinks about it, but if you look at old org charts from 125 years ago, there would be the ‘VP of electricity’ for a power plant,” Hu said.

Today, electricity is everywhere. To Hu, that’s the goal of AI literacy—avoiding friction.

“You don’t want to have to have people thinking really hard, ‘Oh, how do I use this?’ because that’s friction,” Hu said. “It should just be there.”

Take it easy. As with any expansion of technological capabilities, however, there’s a security side. For Rapid7 researcher Deral Heiland, that’s a reason to ensure there are controls on identity access management. Attackers can use devices as springboards to access internal systems, a dangerous use of IoT connected hardware.

“We’re familiar with prompt injection-style attacks,” Heiland said. “That’s been going on for a while, getting more and more mature—but here we’re dealing with devices that appear machine-to-machine, there’s no console, there’s no human interaction, per se, it’s all trusted machines at that point.”