← 返回

2026.10.06 | TB | Networking | Lumen Digital & AI Upsell Potential

原标题：Lumen – Are Digital & AI Initiatives Going to Lead to Upsells?

#networking #datacenter

Third Bridge · 2026-10-06

Analyst-led • 23 Sep 2026

Companies: T US, LUMN US, CCOI US, MSFT US, VZ US

Tim Acker (TA) Operating Partner at BlueThread LLC

John Conca (JC) — Third Bridge Sector Analyst

Agenda

Key Insights

Introduction

JC [00:00:02]: Welcome to Third Bridge's Interview entitled Lumen – Are Digital & AI Initiatives Going to Lead to Upsells? My name is John Conca, and I will be facilitating today's Interview with Mr Tim Acker, former VP and Channel Chief at Lumen Technologies.

Tim, before we start today's Interview, please state I agree or I disagree to the following statement: You understand the definition of material non-public information and agree not to disclose any such information, or any other information which is confidential, during this Interview.

TA [00:00:27]: I agree.

JC [00:00:27]: Could you give a brief introduction to your background?

TA [00:00:35]: Sure. As you noted at the top, I'm the former Vice President of Channel Sales and Channel Chief at Lumen Technologies. I've had about 30 years in technology, telecom and software channels and direct sales from a go-to-market perspective, across the supply chain. Familiar with channels and routes to market in each of those market segments.

Questions

00:01:01] The strategic telecom transition effort has been going on for a few years now. Where do you think the market is going? How is Lumen catching up?

TA [00:01:32]: Yes. I think you hear the company and then broadly the telecom industry in general, talking about transitioning to what Lumen refers to as strategic services or strategic-based telecom or very specifically Lumen refers to as cloudifying telecom. The general idea here is moving from legacy analogue-based networks and traditional routes of connectivity, voice and data, by the way, to more modern stacks, whether those are cloud-based, whether those are newer technologies for actual transport itself, whether it's the use of advanced photonics for speeding up data connectivity, etc. Broadly speaking, for terrestrial telecom, that's one of the biggest objectives that each of them have. Each telecom is a little bit different by market segment. As you know, some of the larger players in the US like an AT&T or Verizon have both a consumer and an enterprise/commercial business. Lumen with its recent divestitures is far more focused on commercial and enterprise than on consumer. Generically speaking, that's the general trend. I think the last thing is going from just providing connectivity, voice and data services is how are they addressing advanced services over the network to various constituents, whether that's to the hyperscalers, whether that's to government institutions, whether that's to classic commercial customers to SMB, the ability to monetise additional services of those networks is becoming increasingly important as the telecom business becomes even more competitive.

00:03:16] Do you think Lumen's or the general telecom industry's primary customer mix is shifting? Historically, it was large enterprises and corporates, but hyperscalers have become their own category.

TA [00:03:56]: Yes. No, I think it is changing. I think that's very natural and let me take you through the cloud evolution. I think it is a great example, and it sets up well for what we're experiencing in AI. As more workloads have moved to the cloud, the hyperscalers today were, of course, many of them at least were the original cloud-based arms dealers, whether it was through SaaS applications or PaaS applications or Infrastructure-as-a-Service. I think that continues on a much more grand scale and you introduce new people and new players into that mix when it relates to AI. I think the move to put more applications off-prem and into the cloud has created a much larger or much newer, I should say, market segment. It's growing very rapidly. If you think about the hyperscalers as a category, they're a bit unique. Of course, as you'd expect, instead of selling to the Fortune 500, 500 times individually, you may be selling to the hyperscalers who have already done the work to bring the Fortune 500 in and now you're connecting applications between data centres and the cloud and on-prem. What I'd tell you is, I think it's both a transition of the market in terms of where applications are being hosted and driven and actually nested.

Then, of course, you can't help but notice the new applications that are actually being created that we never had before. AI, robotics, all the standard acronyms and buzzwords that we always use, but I do think that's creating a whole new class of demand. Of course, that's generated at the hyperscaler level. The last thing I'd say is I think that the selling motion to the hyperscalers is far different. Their needs are much bigger, faster. As you know, the telecoms can't build fast enough or light up fast enough or get enough equipment in fast enough to supply the insatiable demand there. Sorry for the long-winded answer, but I wanted to be complete, I think the question that you asked.

00:06:10] Is the sales motion being different simply a scale issue? On Lumen's Q2 2026 earnings call, Lumen's President and CEO, Kate Johnson, mentioned how Lumen is no longer just dealing with procurement and they're "knocking on the door of the CIO and CTO." Why do you think things are different, using that as a data point? Is it because Lumen is bigger, or is it fundamentally different in other ways?

TA [00:06:55]: I think it's both, the bigger is an obvious answer. Let me take you through both the hyperscaler and then maybe again, I want to take from what Kate said in the quote specifically, but I do think that the change in how you sell to the enterprise changes as well. Let me start with hyperscaler. I think that's very much a pull-based mechanism rather than a push-based mechanism. Very classically, as you know, in enterprise, more often than not, telecoms are pushing services, capabilities, bandwidth, new services into enterprises. Oftentimes, it is selling them, whereas I do believe on the hyperscaler side, it's actually fulfilling demand that the hyperscalers are creating and desperately trying to fill. What I'd say is more often than not, I think the industry is trying to keep up with that insatiable demand that I spoke about earlier as it relates to hyperscalers, where they're standing up buildings, where they're tilting up, where they're dropping CPUs or GPUs. Ultimately, I think the carriers continue to run behind just because I think the need outstrips capacity. Whether that's putting more fibre through same conduit, replacing existing capital equipment in locations or whether that's completely new routes between cities or regions that just have never existed.

Who would have known that fibre in West Texas would have been a big business 10 years ago. While many of the carriers have lit up a lot of dark fibre that has existed for 20-some years that we all remember very firmly in the original infrastructure build, now we've gotten to a relative level of saturation in some of those routes. Now we're creating all net new ones because of the need to put compute closer to areas centres of power, for example. I think it's very unique, and I call that more of the pull by the hyperscalers. On the enterprise side, I think there is more pull from enterprise in general. I think to your point, it is not just selling to procurement. Telecom and data services become less of a commodity. CIOs are expecting more advanced services that are unique between telecom providers. Anybody could sell you a circuit with X amount of capacity and X amount of latency, and that's very easy to put a procurement person in front of and get the best price. As CIOs are demanding things like capabilities at the edge, running advanced services over networks, obviously, new requirements for latency.

I think very importantly, CIOs having to really respond to more aligned business executives who are wanting more automation, more capabilities from their networks. That's unique. I don't think that is what it was 10 or 20 years ago when it was just dumb pipes. These have become very smart pipes. The business and the CIO have to work much more closely together because I think you're talking about how do you use the network to affect business outcomes, not how do you use the network to connect a multisite location. Let me stop there for a second and see if that made sense, or if you need me to elaborate there.

00:10:25] Would a move away from dumb pipes, which has been the old colloquialism for a long time, be at odds with the demand we're seeing for dark fibre and waves? Hyperscalers and even some big corporates are looking to stand up their own networks and own more of it. What about some of the services on top of it?

TA [00:11:24]: I think back to hyperscalers and let's say, neoclouds in that class of customer. Clearly, what they need is speed and fat pipes, so latency and (inaudible 11.39). I think the arms race is there to the point where there's never enough I think there are examples of where hyperscalers are building their own infrastructure. By and large, I don't think that you've seen a go-it-alone strategy from any hyperscaler that has been meaningful enough to displace the assets, let's say, of Lumen. I'm just going to pick a competitor, Lumen and AT&T, Zayo, whomever. What I'd say is, at some point, you're getting to the point of saturation of the size of the network that exists today. I think you're seeing Lumen respond, and you mentioned this, I believe, in what you said we wanted to talk about today, but in terms of intercity and metro fibre investments from not only Lumen, but many others. What I'd say is I think the hyperscalers have realised the cost that they can be prohibitive. I don't think you're seeing a mass build-out.

I think you're seeing the telco step up in such a way to both deploy what they have in the ground and then come up with very creative ways to use both advanced technologies as well as existing conduit assets to help those hyperscalers. Now inevitably, they're still doing some of their own work, especially if you think about undersea. I don't think there's this mass go-it-alone strategy. I don't know if that's cost-effective for the hyperscalers in just the way that you're thinking in the way that you see them leaning on neoclouds vs trying to build every data centre themselves. I think there's a balance to strike there. I'm not certainly an expert in the economics of go it alone vs buy capacity from somebody else. I think Kate and other statements around PCF deals, how the hyperscalers are picking up dark fibre probably stands alone. Back to the enterprise, though, let me go there because I think it's super important. There are services now and there's capabilities in the network that did not exist even five years ago.

I'll point to Lumen because I'm most familiar with it. If you think about NaaS, it's a good example of that is an on-demand metered-based often times application-specific network technology that gives an enterprise a lot of flexibility to turn up and turn down on demand vs creating another port, a long provisioning cycle and having a single-use port that all it does is one thing. I think that that is where you're starting to see more CIOs and more executives in general, realise that they can tune networks with the technology that Lumen is partnering with others to build. I'll stop there for a second, but I think there is a deeper story than dumb pipes. It's about adoption, and it's about making more of those services available as this technology matures. I give Lumen a lot of credit for the maturity it's shown in developing its NaaS technology and really getting out in front of it.

00:15:09] What specific services might be the real demand drivers as we move from old-school TDM [time division multiplexing] products? NaaS [network-as-a-service] could be a pillar. What's outside of that?

TA [00:15:58]: I'll give some examples that I think are relevant. One is I do think that the cybersecurity services over the network that Lumen and its peers are deploying. Again, I'm most familiar with Lumen having recent experience, certainly adds a ton of value. Whether it's classic DDoS, which some would argue is a commodity, but I actually think it's an incredibly important part of what they provide. There are some great examples of how Lumen has leveraged its investments in Black Lotus Labs as an example has productised security services over the network that has actually made those networks much more efficient and has helped CIOs defend the edge of the network more efficiently than just relying on the infrastructure vendors that sit at the edge of the network. In the example of a DDoS or in the example of network-based and I mean operator network-based firewall technologies, if they can shrink the surface area of the threat, then CIO's budget can go toward edge-based cybersecurity and client-based cybersecurity initiatives and spend, frankly, that's much more efficient. I'll use that as one example.

You mentioned enterprise voice. I think it's another good one that I think that the carriers have done a decent job with. Often times, it's partnering with third-party technology companies to make the provisioning of those services over their networks and the routing and capabilities much more efficient. I'd point to some of the work that telcos have done with companies like Microsoft and other cloud-based technologies around voice to make those, again, much more efficient in the cloud and then leverage some of the advanced capabilities that Lumen has built. Then I think there'll be a new class of services. I know that Lumen and others are working on those where they can deliver software capabilities through the network, through the cloud, of course, but through the network to a customer. Think about disaster recovery and that use case application. You need the fattest pipe you can have only at a time when you have an issue. You don't need it turned on all of the rest of the time.

If I think about a use case like that and NaaS, it's another great example of you can run multiple services, turn them up and down over a single port as you need them on demand. Frankly, in legacy networks where every use case or every application had its own port, that would be horrifically expensive. Number of apps, times number of ports, times bandwidth, that's just not super efficient. I think those are some examples, and I might point you to some of the other public examples that Lumen and others are giving around how metered-based networking and that what they're referring to as digitally-enabled networks are actually changing services that they can deliver.

00:19:25] How do you think Lumen customers' NaaS adoption is doing? Is there still a legacy inertia holding on to traditional networking methods? Could it be a time issue? Are we waiting on other next-level higher-tier services that will ultimately force people to drive to NaaS or consumption-based models?

TA [00:20:22]: I think it's a confluence, I'd say that the progress of and I believe they passed 3,000 customers. It was the most recent that I had seen in NaaS really specifically. What's interesting about that, and these are very public examples that I'm sure you can check out. These are not just SMBs that are trying something. These are significant enterprises in areas like entertainment and financial services and other significant enterprises that are actually adopting this. I think they've had some wins in distributed networks, too. What may look like one customer may be a customer with many, many, many locations. If you think about a distributed network, this is another great example of the ability to dial up and dial down as you need. Think about retail, 12 hours a day, that store is closed. Your ability to scale that while the store is open or while business is being done, super important. Back to your deliberate question is I think there is a bit of an enablement gap and a bit of an education gap, number one. This idea that we've sold internet access based on a pipe size forever. Think about processors. You used to buy a PC based on the gigahertz, speed of the processor and the size of memory.

I mean, you still do that to a degree, but not so much anymore. You're really looking at other capabilities. I think that's happening in telecom as well. Old habits die hard. I do think as with that transition of going to consumption-based cloud economics, I think enterprises are still wrapping their head around consumption-based connectivity economics. The carriers are all trying to do a good job of helping people understand how do I avoid runaway bills and how do I budget for that? That's one. I think the second thing is, some players are more advanced in this than others. If you think about it, in order to deliver that capability end-to-end today, you need to have, I'm going to use a simple example. If you have a manufacturing company and you have a facility in Los Angeles and a facility in Pittsburgh, in the past, you needed to have local carrier footprint on both ends of that connection in order to make a NaaS product work between those sites. With off-net capabilities that Lumen or others are building, you'd be able to connect those sites even if the local provider was not that provider, let's say, in Los Angeles, it's a different one in Pittsburgh. That's relatively new.

I think Lumen and other companies are trying to educate the market on that and realise that the footprint has gotten a lot bigger. Think about replacement of the millions of connection points that they have with this new capability. Then the last thing I'd tell you is, that one of the things that probably is unappreciated is that this technology is also important from data centre to prem, data centre to cloud as much as it is, let's say, prem to cloud, prem to data centre, etc. My point there is that there's more applications as that bandwidth and the capabilities rise, there are more points at which a NaaS can be adopted. I think it's a maturation of the technologies. Clearly, Lumen has made some investments there that you know about. I think others are making those investments. There are others that I think are trying to fast follow Lumen, who I think has been a leader in this space.

00:24:14] Lumen has taken some initiatives on the off-prem dilemma. It released the Project Berkeley device six months ago. Could that be the ultimate enabler here? How much difference could that make? The unlock here is the inability to turn up NaaS on an off-net circuit, which was previously capable. That was probably slowing adoption.

TA [00:25:24]: No, it's a fair question. I think it's a big difference maker. I would also say that building the digital infrastructure around provisioning to make NaaS even more efficient for any telco is different. If you think about it, the way that you provision telco services is very classic today. You place an order, circuit gets provisioned on the network. That needs to get then handed off to a local loop provider who's got its own set of SLAs. The time lag alone is a big issue. Replacing that with a more digital experience, which is what I think the carriers are very focused on to make that model more efficient is one piece. There's inevitably how do you provision that. I think back to your question around the technology and what's the gating factor around, let's say, an off-net or moving more locations to be available for services like this is, look, the economic incentives for somebody that's far ahead in this market like Lumen vs some people that don't necessarily have this capability. Those that don't have the capability are not in a big hurry to provide digital enablement to turn up circuits for a local competitor to provide attractive tariff rates on local loops. Those things are still necessary, I think, to make that work.

Particularly, if you think about the scale-up, scale-down nature of a type of product, if local tariff rates on loops are cost prohibitive from that receiving vendor, then the economics get a little challenging. I think it's a combination of those things that will certainly make that better. I do think education is probably one of the biggest ones that in my own experience, getting people's heads wrapped around this idea that you could actually save over time by going to a meter-based service is a bit of a difficult sell. I think there's a little bit of apprehension. I think that that's a big part of what the carriers need to do. Undoubtedly, that off-net capability, products like Berkeley as an example, certainly help that. I think there's a broader story of Berkeley, frankly, but that's certainly something to keep in mind.

00:27:44] Lumen is trying to modernise the business. The NaaS ports and other capacities' prices are interesting. Could the total revenue base realistically grow through this specific consumption model? Why do you think Lumen is diving into this headfirst?

TA [00:28:55]: I think there's a couple of factors at play. Again, I can't really speak to their current business strategy and pricing dynamics really specifically. Let me give you a more, I'll call it, generic version of how the carriers see this. I'm going to use round numbers, these are not representative numbers. I'm just using them as a round number to give you an idea. If I put in a USD 1,000 NaaS circuit, and just assume it just stays at a USD 1,000 rate, it is entirely possible that over time, you deprecate other circuits. It's also more likely that people are adding net new services and not actually taking out services. Because on average, the average site for a company has one internet connection. Maybe there's a dedicated circuit that actually handles something like MPLS or VPN or what have you. More often than not, you get a single circuit. Natural laws of consumption and natural laws of bandwidth suggest that over time, consumption will go up. Today, that's a very painful process to do something like an upgrade at the edge and to incrementally sell more connectivity. It usually requires a circuit replacement. It could involve a competitor coming in there. There's a number of ways that that could actually manifest.

As more services like, let's say, voice or cybersecurity or managed backup or other video applications get added to that circuit and their ability to turn that circuit up almost like a dial exists, you do have some pretty simple math that's price times quantity and if the quantity is the amount of bandwidth that they're consuming, you naturally get some lift there. The second thing is if they can partner with companies in the ecosystem that provide those services, Lumen's ability to deliver those extra applications over the network, whether through a centralised marketplace or in cooperation with one of those providers, there's actually an economic moat that they can create as well. Instead of you going out and establishing, let's say, a separate relationship with a backup services provider, you could order that on demand through your network provider. I think that's another example of a monetisation of network services that today, the carrier doesn't really participate in.

Dumb pipe goes in, then you call, I don't know, they have Microsoft and you get your services from Microsoft. I think that there's an opportunity there for them to provide both their own first-party services and then make that third-party services over the top of the network, both more efficient from a procurement perspective, but then also make that more efficient from a delivery perspective. Companies like Lumen and others, I believe, would like to monetise that opportunity.

00:32:22] How might the partner ecosystem evolve if the network is going the way we're discussing?

TA [00:33:13]: I think that the carriers today, I'll use the hyperscaler or the cloud service provider analogy. If you think about marketplaces, the AWSs and the Azure, etc, that's become a great corollary, I think, to what potentially the telcos would like to offer, either in cooperation with those hyperscalers or those cloud providers or perhaps in competition. Again, I couldn't really comment on their strategy as to whether they see them as friend or foe. I think it's probably co-opetition at some level, too. I think that the powerful part, particularly if you think about… I'm going to use SMB, great example. The average small business probably doesn't have a hyperscaler relationship, and they're not committed to a cloud spend. Boy, it's really efficient if I could get my disaster recovery and whatever other services on the same bill that I get my connectivity through kind of the classic cell phone example, too. Can I get my Netflix, I suppose, or free in those cases from my cell phone provider. I do think that the carriers provide a very efficient distribution mechanism, billing mechanism, customer relationship mechanism. I think if they are able to more effectively monetise that, which I think they've attempted to do with first-party services, and that's been difficult.

The second thing is also there's a need for those services to be delivered technically more efficiently as well. Written and deployed across those networks more efficiently. I think that those relationships between telcos and ecosystem providers, think software providers is a great example. Those are very important to both parties, both from a network efficiency perspective, but then also from a user experience perspective. I think that those ecosystem relationships, if you're an ISV or you're a large hyperscaler and a telco is already billing some 10 million, 20 million, 50 million customers, that's a great route to market. I think that you'll see and have seen in the past, those telcos leverage that as a distribution channel and now I think can actually do that more efficiently. Just real quick, in the past, that's been painful. If I wanted to buy something on my carrier bill, that was like an act of Congress and 16 extra contracts and I had to sign separate documents. If you think about a digital experience and if the carriers can put in a digital experience around their own first-party services and then create a digital experience for third-party services, that becomes pretty compelling for businesses.

JC [00:36:39]: How likely do you think Lumen or other major telcos or carriers are to be successful at this?

TA [00:37:19]: That is the USD 100bn question. Listen, I think that here's the good news. The need for the telcos to change their own digital capabilities to sustain their own business is pushing them this way. What I'd say in the past is even offering first-party services on top of the network in the past was painful because of network architecture, their own architecture. The idea of cloudifying telecom, which Kate talks about, you heard AT&T obviously has rebuilt significant portions of their network in the cloud. I think inevitably, they're modernising their own stack, which they never had in the past, and they have to do it for themselves anyway. I think the underlying plumbing is there or I should say, is being built maybe more accurately. I think that that transition is more of a possibility now than ever before. In the past, it was we can potentially throw bodies at the problem, and that just never worked, and it was a terrible experience. If you can simply, "Go and get an Amazon-like experience to acquire something from a third party over a dumb pipe," you would do that before you request a 35-page contract to get a really simple Microsoft Office 365 licence. I think that their capabilities that they have to build in themselves is going to help drive some of that.

00:38:55] What might be Lumen's biggest competitive threat? Hyperscalers are building decent parts of their network with the PCF [private connectivity fibre] deals. AT&T recently mentioned building a lot of its network in the cloud. Verizon and AT&T are taking a backseat here, and Zayo and Lumen are capturing a lot of the new hyperscale, AI-driven demand. How do you think of the shift over the last couple of months?

TA [00:40:15]: To answer your Lumen question directly, I think the competition stack has become much more simple. If you sell the majority of your consumer fibre assets off, and now that is part of somebody else's organisation, your need to service and make capital investments in two very different businesses certainly allows them to refocus. Obviously, you know that story. I won't belabour that. I'd say it's still a little bit product dependent. I think now with the deprecation of enterprise voice services that's very public, the fact that the consumer business is what it is, and it's far smaller, if not just relegated to certain regulatory areas where legacy CenturyLink would have had obligations to continue to fulfil. Now their focus can really be on enterprise and B2B and government for that matter, I would say. I'd say that the competition stack has gotten a lot smaller. Back to your question in the class of competitors. I think about the Zayo and the Cogents of the world. Classic competitors in the waves and fibre business where they have assets. Lumen is a largely domestic player. That has simplified their stack as well in terms of competition. I think that there are some newer areas of competition that they are very keen on competing in the Megaports of the world, which more closely follow that NaaS-based market.

I think it's actually simplified the the competition. It's certainly intense, I of course, wouldn't rule out the AT&Ts or the Verizons or frankly, the Comcasts, Charters of the world who have assets. Sometimes they're friends, sometimes they're foes. I'd tell you that the competition stack is thinner from a numbers perspective, I think more intense from the race to build and supply. Back to your question around transformation, I think one of the biggest things, and I think Lumen has been very much on record about this. I don't think I'm speaking out of turn, but the ability to thin out the number of different services and focus on higher-value strategic services has been incredibly important to their transformation. I think you're starting to see at times painful, but meaningful progress on focusing on less things and more deliberately on future winners.

JC [00:43:07]: Doesn't that contradict the added complexities of networking and getting this connectivity? In 2025, Lumen mentioned simplifying the business. Shouldn't you offer more services in this increasingly complex and changing telecom world? This seems like a major inflexion.

TA [00:44:44]: Yes. Listen, I don't need to guess. I know that it is one thing to say, "Hey, you know what, we've got 10,000 products, we're going to get it to five." Then you start to peel that onion back and you realise, "Oh my gosh, if I get rid of these 30, I actually have an enterprise customer that's got one of these other ones that I really need to keep their business." At times, there's trade-offs that weren't incredibly obvious and damaging that relationship, because, again, when you do that, especially in areas like enterprise or government, you put the complexity back on the customer and now the customer is incurring costs for your own zealousness to retire products and not giving them enough runway. I think the wireless carriers are a great example of this. We talked about removing 2G from wireless and 3G from wireless networks 15 years ago. It's still alive. It's still painful in certain areas. I think that inevitably, there was an expectation that Lumen and others, frankly, could move a lot faster. I was not there during recent discussion around voice services and decisions that were there and certainly was not in any of those rooms.

You can imagine there's probably a great trade-off to say we've provided these for 30 years, and we are probably going to upset some customers. There's natural churn in these products, and we have to make some hard decisions very much like I think that Lumen made substantially around the divestiture of their consumer assets. That was a great business, but it was what do you believe is the future of the company? In the case of the consumer fibre business, it was what is going to be best for those consumers and best for that side of the business. I think the leadership team made a decision that being with AT&T would be the best thing for both. I think that's a decision that I think should be applauded and a very tough one because it was a good business, too. I think it's not a binary decision. I think over time, as they unpack the network and then unpack some of the challenges and actually turning off services, what that would involve and the risk that they would put customers in their own business at. As you unpack that onion or unpeel the onion, you find out more harsh realities.

00:47:18] Strategic revenue now makes up 53% of Lumen's revenue contribution. How are you thinking about the growth runway for strategic revenue? It's worth noting that the company's strategic revenue grew at 14.1% in Q2 2026, which was significantly higher than Q1 2026's 9.4%. Could double-digit growth in strategic revenues be a sustainable long-term forecast? It ultimately goes back to the adoption curve question.

TA [00:48:46]: Yes, I don't know how to answer the question directly because I think there's two factors that are impacting it that I don't know what the end of the runway looks like. Here just very plainly is many times, strategic revenue is replacing legacy revenue. You see churn in one is being replaced by the other. Then quite honestly, strategic revenue includes things like NaaS and includes things like dark fibre, and it includes things like PCF. I think if hyperscaler demand, as an example, remains as robust as it is, that could keep going for some time. Forgive me, I'm probably not as big of an expert on the hyperscaler and how far this train keeps moving. I think everybody is asking that question. I think you see a good portion there. Look, I do think that they've done a good job in creating the correlation between, "Hey, here's this legacy TDM. Here's a viable alternative. Where there is a viable alternative, there are some products that just are not replaceable." My suspicion is that between hyperscaler growth which, of course, includes PCF and change out of legacy products, I think that's contributing to it. I'd caution if you look at the hurdle rate of competitors who are potentially more mature in digital technologies, I don't know that they're accelerating at those rates.

00:50:25] We discussed differences in selling motions and how hyperscalers are becoming their own class. Do you consider it a prerequisite that the hyperscalers spend on that for strategic growth? Massive CAPEX is punishing some of Lumen's stocks and every earnings report. Could this growth be hyperscale-driven or linked to AI?

TA [00:51:40]: I think it's highly correlated. What I'd say is and this is not a Lumen discussion. I think this is a broad telecom discussion. I don't know that applications post AI data centre growth are accelerating at a rate that would fill the gap. What I mean by that is inferencing at the edge or edge computing or moving from training and query and agentic to industry-based AI applications. I don't think if the slowdown were to happen tomorrow, I don't believe there is enough in those other technologies or those other use case applications to fill that void is what I'd tell you. I just don't think that we're mature enough and we're there yet.

00:52:37] Is there anything else you want to mention that might be important for the investment community looking at Lumen? What do you think people still misunderstand?

TA [00:52:58]: I think we started to touch on it. I think that for some of your work, thinking about routes to market in general, there is through hyperscaler, there's through eco, there is eco back through, let's say, Lumen or telco. I also think that the telcos in general are thinking very deeply about partnerships, channel routes to market and why that is probably going to continue to be an efficient way if they manage it correctly, to continue to grow their revenues without growing their SG&A at the clip it has grown in proportion to the revenue in the past. I think that's the only other thing I'd say is the ability for them to really think deeply about go-to-market customer acquisition and adding services going from just acquisition to expand and adoption, that's critical for every one of their businesses. Other than that, I really enjoyed our time.

JC [00:54:09]: What might be the single biggest risk if this transformation stalls? What would cause it?

TA [00:54:35]: Yes. I mean that's the sexy top headline. I do think equipment availability, the ability to deploy the network fast enough, availability of fibre. Infrastructure availability is a big one. I know the company has been very vocal about this is that they would like to get their hands on more equipment to help with more deployment. I think that's the other big risk outside of the macro AI and corporate spending that I think drives a lot of the telco economics.

JC [00:55:10]: Thank you so much for your time and your commentary today.

TA [00:55:15]: Thank you.

Transcription ends at 00:55:17 of the recorded material