Verizon Business is in the process of scaling its artificial intelligence (AI) capabilities to better support the management of what is becoming an increasingly complicated mixture of telecommunication services.

Wael Faheem, SVP and CIO of Verizon Business Group’s Global Technology Solutions operations, told SDxCentral that new network technologies like mobile edge compute (multi-access edge computing (MEC) and private networks can add a layer of management complexity to already overburdened enterprise IT teams, which can be supported by different AI models.

“You can maybe let the network do some of the AI, but also you can give the implementers on the customer side a step-by-step in a way using AI on how to set up network, how to manage their network, how to engage with us if they are opening a ticket on their side,” Faheem said. “It doesn’t have to be AI 100%, but just giving them a way that’s easy for them to open tickets in just the updates and then we can do the trouble ticketing behind the scenes.”

Verizon’s network angle on this opportunity is also helped by it having deep understanding of its own network architecture and the data flowing across that network as well as an open interest in helping an enterprise manage their own legacy network architecture. This in turn can be helped by AI.

“We have to do this for every single customer,” Faheem said. “Watch their network, scale our AI models to predict what's happening in their network, what could go wrong in their network, and then tell them, ‘hey, you know what? You might want to click this button. You might want to reset your network.’ We might tell them, ‘I see something wrong in your network. I'm going to be sending a technician to your premise to fix that.’ They might not even know that there's something about to fail.”

Verizon steers network AI toward greater simplicity While the process sounds complicated, Faheem added that the key for making this work in the real world is to make it as simple as possible for those IT teams. For Verizon, the answer to this is abstracting that complication away from the network.

“It doesn't matter if it's 5G. It doesn't matter if it's riding on fixed-wireless access [FWA] or wireline network. … All that is access,” Faheem said. “We're going to give the customer connectivity. It could be any type of connectivity, and in our systems we have to also provide that type of service, that type of abstraction. And then on top of it, we’re going to give you connectivity, and then we're going to tell you what are the available services, what are the available solutions on top of that connectivity.”

Those services and solutions can be built on MEC or private networks that integrate with on-premises Wi-Fi connectivity or go so far as a fully managed SD-WAN service similar to what Verizon Business recently accomplished for the Commonwealth of Virginia. That deal included a 1,000-site managed SD-WAN deployment that could support public and private networks.

“The technology folks on the customer side, they don’t want to learn something new. They just expect us to make it so simple for them and tell them what they are going to get,” Faheem said. “You're going to have connectivity on your premises. It doesn't matter that it's Wi Fi versus mobile access directly as long as we give them that access, they will be fine.”

This leads back to Verizon Business’ increased focus on scaling its AI capabilities to help ease the deployment and management of these network systems.

“We've done a lot of great work that I'm very proud of across Verizon in that space. Now, it’s how can we scale that? How can we infuse everything that we're doing in the systems and the network and for the customer experience in all our products across the world? I would say that's our top priority here,” Faheem added.