This discussion from Mobile World Congress 2026 in Barcelona examines telecom strategies for artificial intelligence and the hyper-converged edge. Zeus Kerravala of ZK Research, founder and principal analyst, provides analysis on how service providers leverage AI, edge architectures and partnerships to capture new revenue.
Kerravala draws on decades of telecom and networking analysis in this theCUBE Research discussion with host John Furrier of theCUBE and on-floor reporting by Dave Vellante of theCUBE. They explain how software-defined networks, modern silicon and hyper-converged edge architectures converge with AI to reshape service provider footprints, partnerships with hyperscalers and operational models required to deliver industry-specific solutions.
Kerravala identifies agentic ops as a factor to dramatically reduce operational costs and positions the hyper-converged edge as a pivotal revenue opportunity for telecommunications providers. They emphasize the importance of real customer deployments and verticalized use cases, and the theCUBE hosts underscore trust, compliance and demonstrable return on investment as prerequisites for broader market adoption and investor confidence. Viewers gain insights into practical deployment considerations and partnership models for monetizing edge and AI initiatives.
Forgot Password
Almost there!
We just sent you a verification email. Please verify your account to gain access to
MWC Barcelona 2026. If you don’t think you received an email check your
spam folder.
In order to sign in, enter the email address you used to registered for the event. Once completed, you will receive an email with a verification link. Open the link to automatically sign into the site.
Register for MWC Barcelona 2026
Please fill out the information below. You will receive an email with a verification link confirming your registration. Click the link to automatically sign into the site.
You’re almost there!
We just sent you a verification email. Please click the verification button in the email. Once your email address is verified, you will have full access to all event content for MWC Barcelona 2026.
I want my badge and interests to be visible to all attendees.
Checking this box will display your presense on the attendees list, view your profile and allow other attendees to contact you via 1-1 chat. Read the Privacy Policy. At any time, you can choose to disable this preference.
Select your Interests!
add
Upload your photo
Uploading..
OR
Connect via Twitter
Connect via Linkedin
EDIT PASSWORD
Share
Forgot Password
Almost there!
We just sent you a verification email. Please verify your account to gain access to
MWC Barcelona 2026. If you don’t think you received an email check your
spam folder.
In order to sign in, enter the email address you used to registered for the event. Once completed, you will receive an email with a verification link. Open the link to automatically sign into the site.
Sign in to gain access to MWC Barcelona 2026
Please sign in with LinkedIn to continue to MWC Barcelona 2026. Signing in with LinkedIn ensures a professional environment.
Are you sure you want to remove access rights for this user?
Details
Manage Access
email address
Community Invitation
Zeus Kerravala, ZK Research
John Furrier hosts a conversation with Zeus Kerravala, Founder and Principal Analyst at ZK Research as part of theCUBE’s coverage of MWC26 from Barcelona, Spain
In this interview from MWC Barcelona 2026, Zeus Kerravala, founder and principal analyst at ZK Research, joins theCUBE’s John Furrier to discuss how the telecom industry is navigating the shift from connectivity-centric models to the hyper-converged edge. Kerravala breaks down why this era is different for service providers, noting that the transition to software-based 6G and AI networking eliminates the need to build entirely new physical infrastructures for every technology shift. He examines the rise of "agentic ops" and the potential for telcos to finally...Read more
exploreKeep Exploring
What is your assessment of the event so far (day one), particularly regarding the prominence of AI, hyper‑converged edge, quantum/intelligent infrastructure, and the implications for telecom/service providers?add
How is this moment different for telecommunications companies, and how do software-based networking, 6G/AI networking, and partnerships with hyperscalers affect their ability to capture value?add
How can I determine whether a networking/cloud product or vendor offering is truly AI-ready and actually delivering real-world benefits to customers?add
Where should agentic AI agents be applied, specifically in the telecom industry? Beyond cost optimization, are there revenue-side opportunities?add
>> Welcome back everyone. Here's theCUBE's live coverage here. I'm John Furrier. You have Dave Vellante out on the assignment on the show floor, getting all the stories here at MWC 2026, three days of wall-to-wall coverage. Zeus Kerravala's here. He is the founder and principal analyst ZK Research, CUBE alumni, CUBE contributor. Zeus, as always, great to have you on theCUBE, and I'm expecting a blog post on SiliconANGLE shortly.>> Oh, yeah. There'll be a few.>> Your patented move, come on theCUBE and get a post up. Rob Hofe loves those stories. You're really going to do a great job of synthesizing what's happening at the show. First I want to ask you, this reminds me of super computing five years ago when it turned into from an HPC show to an AI show. Full-blown AI, a lot of enterprise, a lot of hyper convergence discussions around the edge of the network. Talk about quantum and intelligent infrastructure. This seems to be the show theme. What is your read so far here on day one?>> Yeah. Well, every show you go to turn into an AI show. And it's interesting with this being a service provider show, John. Where so much of the focus here is helping service providers try and figure out how to do that mythical thing we always talk about, but they never do, which is find new sources of revenue. I don't think there's any question that AI is going to help them lower their operational costs. I think agentic ops is coming, and that's going to make them a lot more efficient. But what I've been looking for is can somebody answer the question, once AI is in the network, do the telcos actually have an opportunity to raise their value prop and create new sources of revenue? And I think they're going to have the opportunity. I'm not so sure a lot of them are set up to do that, but I think that's maybe the sub theme here is ->> It feels different. Because you and I both have joked many times about, I called them, they move as fast as glaciers. And really, not a compliment to the telecom providers. But that was pretty hybrid edge. Cloud was happening. Was everything teed up? It seems like with distributed computing and hybrid, it seems like that's rolling to their front door. They have footprint. Whether you're operating with towers and power and cabinet and conductivity or you're running a central office, a lot of the carriers and operators have infrastructure. They have the coveted energy. I think this is an opportunity. In fact, the headline I'm writing for SiliconANGLE right now, to your point, I want to get your thoughts on it, says "The hyper-converged edge is telecom's next big bet - or its last missed opportunity." This, to me, seems like the final moment of consequence for the telecoms. Because, yeah, they missed cloud. Apple, Google picked up the rents there. And then with the marketplaces, SaaS accomplishes happening now. They have the data, they got the networking, they have the footprints. Yeah, there's an internal network piece that HP and Juniper and Cisco will provide gear towards. But the wireless piece, AI factories coming to the edge, it's a matter of time.>> Yeah. I think what's different this time, and I went to the Nokia event yesterday and their CEO, Justin Hotard, was talking about how when you think of these big moments in networking, right? We had voice networks and data networks and then cloud, and the network is the thing that connected them all. The challenge back then for telcos, John, was that every time we went through these shifts, they had to build an entirely new network. The network that delivered frame relay was different than the network that delivered voice. And that was different than the network that MPLS was going to say. Every technology shift, they had to build a new network. Now, Jensen talks about this all the time. With 6G coming and with AI networking coming, the big difference in networks today is they are software based, not hardware based. So they don't have to go build a whole new network, they can use the infrastructure in place. And doing the innovation either in the cloud by partnering with Google or an Amazon, or they can do it just in software themselves. I think this is their best opportunity to capitalize because the lift isn't nearly as heavy as it has been in the past.>> In terms of their ability to capture->> Yes.... >> the value.>> And the speed which they can work with. If you've got to build a whole new network to go do something, well, that's going to take you a long time. Now, they've got the partnerships in place with the hyperscalers, they got the software models in place, all these equipment vendors are helping them modernize their equipment. And so, they should have the opportunity to move faster than they have in the past.>> What is your take on culture? Because the culture in the legacy side is cell conductivity, revenue per circuit. Now it's managed service, AI services. Now you've got revenue per service. They have to get out of the box mode. Because the old way was, here's a box, monetize the connectivity.>> Well, a lot of the world has moved to outcome based selling, and I don't think the service providers do that all that well. Well, that's not true as a blanket statement. There's some regions in the world where do telecom. And the UAE, Telstra and Australia, BT, they do a good job of that. By and large, it was a group, especially the US based operators, they don't do such a good job of being able to talk about how the impact of connectivity and transformation there is going to help transform businesses. And this is where they rely on the equipment vendor partners to help do that.>> When you look at the show floor, you've talked about it in the past, AI adjacency versus being primary AI native. What's your smell test for AI washing out on the floor? You look at the people, certainly there's a lot of quantum washing right now. But AI specific, AI ready infrastructure, how do you tell the AI washing people who are just hand waving versus the real deal?>> I guess a lot of it is just the age of it, right? If you're taking an older box that was never meant for AI and you're trying to add another layer on top of another software layer, it's probably not AI ready. I think you got to be running modernized silicon at its most basic level. The newest Broadcom, Silicon One, NVIDIA GPUs, whatever, right? But it has to be that. But I think more importantly, are they actually delivering on it right now? There's so much vision out there and so much talk about products that are coming and products that are coming, but a lot of it's not here yet. For me, it just comes down, do you have customers that have deployed it and what are the benefits they're getting from it? There's some. I was in AWS's demo area this morning and they talked about how the... I think Telus is one of the customers, dropped operational costs by 80%, things like that. The case studies are there. There's fewer of them than more of them. But that, to me, is the litmus test. Do you have customers using the products that are gaining an advantage today?>> You mentioned agentic earlier. Where do you see agents working? You mentioned cost optimization. Any on the revenue side? What's your view of agents in the telecom AI world?>> Yeah, that's a good question. And this is an industry that actually could use them. Because historically, when was the last time you called Verizon or Comcast, or one of those companies?>> I remember.>> Yeah. And when you do, you don't get a great experience, right? They have network outages, things like that, but they could easily use agents for customer service as the low hanging fruit. Because when you look at the NPS scores of the service providers, they're generally pretty low when compared to their software and infrastructure peers. If they can just bring agents in to handle a lot of the burstiness that you get when things on your network happen, that is a huge win. I think from a low hanging fruit perspective, it's on the customer service side.>> Remember the days back, go back 15 years ago, the buzzword was the consumerization of IT, and that was an enterprise story. We have a consumerization of access. I want to ask you, with wireless and hyper-converged on the table, what does the enterprises look for right now? Sovereignty over latency? Performance? What's your view on the enterprise coming into this, what is ultimately, in telecom, a consumer business with a siloed enterprise group? Does it come together? Or, what's your take?>> Well, it can. In fact, there are companies out there, Zscaler, Cloudflare, that are pushing this concept of moving away from your legacy WAN to something that's more cafe. You've heard this model, right? Where you treat every user... And look, it's not for every company, but if you're a retailer, insurance company, things like that and you've got a lot of distributed employees, it's a better model. And I will say, one of the big holdbacks there, barriers against that, is Gartner. Gartner does not recognize that as a viable WAN strategy, but I think it is. And I think there's so much connectivity out there, connectivity options for businesses, they got to stop thinking about these... They really have to rethink how I architect my network with the right type of connectivity is there. I don't need to have a traditional router and firewall on every location, maybe I just need connectivity. Consumer connectivity, in a lot of ways, is better than enterprise connectivity. It's faster. It's more reliable in some cases. With a little bit of tweaking, you're going to have a network that's a lot cheaper. I was talking... You know Zuora, their San Mateo base. I was talking to the CIO there. They moved to this model of get rid of your WAN, go to this cafe style, consumer grade broadband. They dropped the cost of their network by 80%. The proof's there.>> All right. I have to ask you, I know you do a lot of reporting on the floor. You mentioned AWS, you mentioned some of the people you were visiting. On the supplier side, on the vendors, what's the vibe? What are you hearing? What's some of the conversations you're having?>> There's a lot of excitement. And clearly everybody recognizes that AI is going to change traffic patterns and that'll create an opportunity to go help companies modernize their network. I think there's a lot of pent-up optimism that we're going to see a massive upgrade, both on the service ledger side, enterprise data center and campus to be able to help build better training clusters, but then also inferencing at the edge. Edge is coming, physical AI's coming, which will drive more edge connectivity, more cellular connectivity, more wifi, things like that. There's a lot of anticipation. I think the capital markets reaction recently though, John, has been very confusing to a lot of public traded companies. Where you see companies put up beaten raises and the stock get crushed. The entire security sector is in a downward trend right now, even though one would think that the more AI I have, the bigger my attack surface gets, the more security I need in more places, right? But I think->> Is that a gap between understanding the value proposition or the value creation piece, or is it more psychological fear of AI bubble?>> I think it's a bit of both. And I think there is a fear of AI bubble. And I think Wall Street's now saying, "Prove it to me." You've talked a lot about how infrastructure spend is coming or refresh is come in, now prove it to me. Let's see this. Because I think as long as enterprise is sitting on the fence and push off product, push off product. Which, let's face it, they are. That creates a lag between what's been said. And I think Wall Street wants some proof now, and it's a fair statement.>> Well, I'm psyched to have you on, good analysis. What's your personal focus now? What's got your attention? What's on your radar?>> Well, I want to see use cases. I, like everybody else, have heard the vision, and I want to go talk to companies that have deployed it. When I go talk to the vendors, I want them to tell me about customers that deployed it, the return they got. More and more, I was at the NRF show this year, I'm going to go to HIMSS this year too. The vertical shows become more interesting. I'm really curious to see if the... I asked G2 this question. I know you had them on this morning. With Cisco, do they need to start thinking about... And the other vendor's putting a retail or healthcare or insurers, or whatever, some kind of industry wrapper around their products to help customers understand how to use these things in their specific industries. That's what I'm looking for.>> Use cases.>> Yeah, your use case.>> And you think the verticals are hot?>> Yes. Well, that's->> They should be.>> Should be hot. AI should be domain specific value.>> Yes.>> In fact, IBM talks a lot about that, the growth being in small language models that are more industry specific. We're going to train everything core screen using LLMs, then we're going to fine tune them using small language models for specific use cases. The other area too, John, I'm sure you've talked to this, everybody you've add on, is the concept of trust. I think this global macro we're in right now, and this started from a conversation at Davos. Every conversation I have has some element of trust to it. Governments don't trust each other. People don't trust their AI systems. We're not sure that we can secure them. My employees are doing things in a shadow AI perspective that I can't trust, so I don't know what data to put in, what AI systems where. There's so much uncertainty out there right now. That does favor the big vendors though, because they've got more trustworthiness, I guess, because they've been around, they've got the channel partners.>> I was talking with the president emeritus at Cohere, Kevin, he was talking about how the compliance work, if you do that work upfront, it's a grind. That's my word, he didn't say grind. He said, if you just take the time to get past the nos, because every risk management department's, "No, no, no, no, no." But you have to just hold the line and just do the work to get them comfortable. And once compliance hits, the agents are more effective, they happen faster. There's almost a playbook emerging where that compliance ground game work->> Measure twice, cut once.... >> is very much in play. And that's a theme I'm hearing a lot more. I'm hearing a lot of, "Well, compliance and governance is critical." Well, that's cool. I get that, but show me. And that's just the use case, just got to grind it out.>> Yeah. I thought the... At Cisco's AI summit, I know you watched that. Matt Garman's presentation I thought really was a good one, the AWS CEO. Where he talked about how if you've got a canyon and you put a board across it, you're going to crawl. But if you've got handrails up, or guardrails, you're going to run. And I think that's what we're looking for right now is we've got the right compliance and security foundation in place. Enterprises now can run to AI, not crawl to AI, and that's good for the whole industry. I just don't think that it's a natural motion for us in tech, John, is to think about security first. We deploy the stuff. My friend fell off the board. Okay, see you later. I never liked him. Maybe we need some guardrails. But historically we deploy stuff and they go, "Oh, crap, it led to all these new problems. Let's go secure it now." But we got to rethink this and do security first.>> Zeus, it's great to have you on. I love chatting with you. Use cases and examples. Show me the money, show me the use case, show me the proof. We are in a market where certainly the hype and has to be matched with reality, as we're trying to do here in theCUBE. I'm John Furrier with theCUBE. Thanks for watching.
