T-Mobile US sees a cloud-based revenue opportunity tied to its future-looking push of bolstering its network with artificial intelligence (AI) to support edge components of its radio access network (RAN) architecture.

CEO Mike Sievert noted during the carrier’s recent investor conference that AI advances and adoption are “unfolding like every other technology innovation of our lifetime in one aspect, which is it starts with simple text and then turns into more immersive experiences later.”

This path aligns with where T-Mobile US is heading with its AI-RAN push that will see the carrier partner with Nvidia, Ericsson, and Nokia on an AI-RAN Innovation Center. That facility will house a focused effort on tying together cloud-based RAN and AI development.

The carrier noted in a recent presentation that the goal is to integrate cloud-based RAN and AI using unified infrastructure that can scale to serve millions of mobile users at once.

“AI-RAN will enable new AI algorithms to unlock the full potential of wireless networks,” T-Mobile US’ presentation noted. “These AI algorithms would be rapidly developed with software-defined RAN, trained on AI data centers, and fine-tuned with physically accurate digital twins. This will lead to dramatic improvements in spectral and energy efficiencies.”

This architecture plan will see operators update their RAN equipment deployed at either each cell site or at a central location that can serve several cell sites. These updates will turn those locations into edge data centers or cloud connections that can in turn better serve AI workloads.

“These kinds of AI workloads will increasingly demand that processing have happen on the device, which we're starting to see but there are obvious limitations there, or on a cloud near the device, and that's a business opportunity that we see in the future,” Sievert said, though speaking at an investor event quickly added that “there are zero-dollars in our plan for this.”

However, Sievert further aspired that “we do see that running third-party AI workloads in our magenta cloud near the customer endpoint may be something that we can add value to the world with. There's nothing in our contract about this because that cloud would be ours, not belonging to our partners, and so what we would be doing is assembling the network technologies to do it in the future.”

That “contract” reference is tied to T-Mobile US’ current work with cloud hyperscalers like Amazon Web Services (AWS), Microsoft Azure, and Google Cloud Platform (GCP), which could also be part of this still evolving process.

“We would be very open minded about partnering if there's a way we can,” Sievert explained. “We're a deep partner with all three, probably deepest with Microsoft Azure, but we're also work deeply with AWS, we work with GCP, those are the scaled providers. And you saw we are working with OpenAI that right now all the workloads are processed in Azure, but they're a close partner. There's lots of flexibility here and if we can add value to our partners and have joint productization and go to market in the future, that's one area to explore. But that's pretty speculative at this point.”

Rivals, partners see AI edge opportunity

Sievert’s comments echoed statements that Verizon CEO Hans Vestberg has made over the years in touting that’s carrier’s multi-access edge compute (MEC) strategy.

Vestberg during a recent investor conference noted that while the market itself around AI and generative AI (genAI) continues to gain momentum, it could still be some time before market dynamics shine on true edge infrastructure.

“The loads are coming,” Vestberg said during the Goldman Sachs Communacopia + Technology Conference. “Probably it’s going to take some time because the majority of genAI today is large language models that you’re training, so they send them way back to the data centers. But as soon as they start doing, like Verizon, it’s an application that you use, then you want it closer to the customer because of transport costs, security, [and] in certain cases, latency.”

Vestberg also boasted of Verizon’s long-standing partnerships with all three of the nation’s largest hyperscalers to boost its edge posture.

“We have processing, compute, storage, power already built across the nation with our mobile edge compute, so I think that no one in the telco world is better prepared than Verizon to be part of the genAI edge compute of the time,” Vestberg said.

Analysts have noted that the hyperscalers are angled to gain a significant share of these edge and AI opportunities.

“With their vast access to data, the hyperscalers will be a natural choice for customers that do not want to move their genAI solutions to a separate compute ecosystem but want to process the data closer to the point of use,” Technology Business Research noted in a recent report.