Keynote: Kenny Gunderman, President and CEO, Uniti Group Inc.

Chip, thank you for your comments. We're thrilled to be here. I've realized I know about 50% of the people in the room, Chip, so I want to rethink opening up questions to the audience. I'd rather not do that. But great to be here among friends in the industry. It's always fun. Super excited to be invited up here today. Please read our forward-looking statement. It'll knock your socks off, I promise, but I have to tell you to do that for the lawyers. You can find copies of this later. We were last asked to give the keynote here back in 2017, right on the heels of our acquisition of Southern Light at the time. There's a table full of former Southern Light employees here today: Eric Daniels, Ren Landers, Kelly McGriff. Kelly's probably passed his card out to everybody already this morning. They are now contributors to Uniti; they have been for seven or eight years. We're very excited about that transaction. Chip mentioned Windstream; we're very excited about that combination. Is Joe Scaria in here somewhere? There he is in the back. He won't stand up, but he raised his hand. Joe runs the wholesale business at Windstream, and we look forward to seven or eight years from now looking back at an equally prosperous and mutually beneficial transaction there. But the lesson is: if you want to be the keynote at Encompass, you have to do an M&A deal. That's what gets you invited up here. Back in 2017, I thought what might be interesting today is to go back and look at what we said in 2017 and compare it to what actually happened, because we did some forward-looking comments. But then I went back and looked at our presentation and thought, nah, maybe we don't want to do that. We weren't right on everything, but there were a few things we were right on. One of those was that we thought emphatically that fiber was going to become the next mission-critical communications infrastructure asset. We said it emphatically, and we were right. We were right about that. Everybody in this room knows that we're in the epicenter of a lot of the themes Chip just mentioned: hyperscaler demand, AI demand, generative AI, waves, fiber to the home, convergence. So couldn't be more excited about where we are in the industry, couldn't be more excited about the fact that fiber is mission-critical and that we've got an expansive fiber network at Uniti. But despite that, when I travel around and talk to investors in the industry—whether it's public, private, strategic, or financial—we at Uniti still get the question a lot: 'Why is the fiber industry better today than it was back in the 2000s?' Let's face it, as an industry in the early days, we made some mistakes, and there are still people who remember that. So we get that question a lot, and frankly, I love the question because we've got a lot of great things to say about why. I wanted to spend a few minutes on that today and maybe have a conversation about it later in Q&A. Number one, and it really boils down to three things: number one, it's just a much more rational competitive landscape. We're not overbuilding as much; there are much more targeted capital investment models today. Number two, there are a lot more of what we call on-ramps—all those last-mile, high-broadband feeders of capacity onto networks. Back in the 2000s, we were building a lot of long-haul fiber, just not a lot of on-ramps. Number three, as a result of those first two, back in the 2000s demand was on the come. We were building things with demand on the come. Today, we are awash in broadband demand. It's terrific and awesome. So I wanted to spend a few minutes talking about each of those topics. I can't really see the slides from here, but I've looked at them. As Chip said, I joined the telecom industry back in 1998, a couple years after the Telecom Act of '96 was signed by fellow Arkansan Bill Clinton. No connection. Trip down memory lane: there was a lot of excitement about the industry, a lot of capital chasing the industry. Everybody wanted to be a CLEC, and there was this huge excitement about the broadband demand that was on the come. Capital was chasing business models, and long-haul business models in particular. So you can see there's a lot of logos on here that are familiar to people in this room, and you can see that the majority of the fiber being built was long-haul in nature, and just a handful of companies were starting to really focus on last mile. Fast forward to today, and many of those long-haul networks—big, beautiful networks, terrific infrastructure—they were just empty 20 years ago. Fast forward to today: much more rational industry. A handful of competitive long-haul providers with networks that are full of capacity, full of traffic. And along the way, there's also been a lot of investment made in fiber to the home, metro fiber, and fiber to buildings. As you can see, the increase of long-haul fiber from 700,000 to 1.4 million, but a massive increase in last-mile fiber. That's the sign of a much more rational industry. And all those logos in the bottom right are not overbuilding; they're not building the same tier-one long-haul routes. They're building unique fiber in unique markets. Largely, there's some overbuilding, but not nearly as much. Much more rational. This is a great page. I love this page. So what is all that fiber being built mean? If you look at the bottom left, back in the early 2000s there were a few thousand fiber-to-the-home customers. Today we have 80 million fiber-to-the-home passings and growing at a substantial rate, because there are still many homes in this country that don't have fiber to the home. And the business model for fiber to the home works today. You can build fiber to the home at a level and have a clear line of sight to penetration and ARPU, and therefore the returns work. Back in the early 2000s, it didn't work. There were still people using dial-up in the early 2000s. Anybody in here remember dial-up? It was awesome and it sucked at the same time. It was all we had. But man, today compare that to a 1-gig fiber connection into the home. There's no comparison. It takes 375 years to download the Library of Congress on a dial-up connection; it takes a matter of hours on a 1-gig fiber home connection. So that bottom-left bar chart shows the number of passings, but it doesn't begin to capture the amount of demand that fiber to the home is going to generate. That's only part of the equation, though. Back in the early 2000s, cell phones were starting to be used, and I remember people saying, 'Eventually we're going to have these single handheld devices where you can check the web, check your email, and talk on the same device.' I remember thinking that's crazy, there's no way. But today, not only do we have cell phones, we've got these supercomputers. We can all go to the moon because these are better than what NASA had to get to the moon in the late '70s. Massive amounts of capacity running through all of these supercomputers, and again, that doesn't work without fiber. The middle bar chart: huge increase in towers and small cells that enable us to use cellular phones and smartphones. None of that stuff works without fiber. In the early days of towers, a lot of people in the room remember Ron Mudre from Tower Cloud. A lot of those initial towers were T1 or even microwave. That doesn't work today. Today you need 1-gig fiber connectivity, 10-gig fiber connectivity, 25-gig right around the corner, and so on. So if you put a demand curve next to the number of cell sites, it's exponentially higher than just the number of actual sites. Again, just tremendous amounts of demand. Number of lit buildings: same thing. Roughly 300,000 lit buildings with fiber in early 2000. Today, Uniti—the new Uniti—is going to have over 800,000 near-net, on-net fiber-fed buildings alone. Just us. Now I don't think that's exactly apples to apples to the 300,000, but close enough. The point is, there's just been a huge increase in getting fiber to enterprises in addition to individual consumers and homes. The social media point: not really an on-ramp, it just shows how many more of us today are using the internet than we were before. You all know that, but I think it's super interesting to show. It's not a communications infrastructure presentation unless there's an up-and-to-the-right demand curve. So here's ours. This one comes from our friends at Cisco. We've used it in some of our investor relations material. I think it's interesting. We could spend time on each of these bars. Digital transformation: that's the phase we're moving out of, according to Cisco, and I agree. I don't call it the digital transformation phase; I kind of call it the COVID phase. Pre-COVID, post-COVID. To me, COVID, as terrible as it was—set aside politics, whether it made sense for us to be separated for social or political reasons, it doesn't matter—we were, and it would not have worked had we not been able to stay connected virtually. I think as a result, COVID has really accelerated the focus on fiber as being mission-critical, and it's really accelerated people's views of all the different things that fiber can bring to us. I think without COVID, AI generative AI would still be five, six, seven years out. But these hyperscalers, who are very smart and have lots of capital, they see COVID, they see the opportunity it brings, frankly they see what it meant to their business models, and now they want more. That's leading us to that far-right phase. Again, massive increases in demand, and these are driven by AI, yes, but it's also driven by a lot of other use cases. I'm not a technologist; I don't understand a lot of these things on the granular level. But as an infrastructure guy, I understand that you're going to need a lot of fiber to support these use cases. A couple weeks ago, Elon Musk rolled out his 'We, Robot' strategy, I think it's what it's called. He showed his autonomous vehicle fleet, and he also rolled out what he's calling the Optimus robot, which basically emulates human beings' voice, actions, dexterity. He predicted that the Optimus robot will be the single biggest product ever sold in the world, and it retails for $30,000. So I thought, okay, there's probably a bunch of us that want one of those. There's probably a bunch of people who would prioritize other things over a robot for a while, at least. But let's say he's 5% right or 10% right. That's a lot of robots that get connected via wireless back to towers that traverse a lot of fiber. That kind of thinking is awesome and terrific, and we've got a lot of really smart people with a lot of capital out there thinking about those use cases on our behalf, because we benefit from that not only as consumers but as communications infrastructure providers. So this page just shows a progression of our customers as infrastructure providers. Back in the early 2000s, there weren't really a lot of customers for fiber. There were wireline customers, there were wireless customers, a few large enterprises were acquiring fiber. Good customers. Many of those businesses are still customers today. But when you fast forward to today, there are a lot more customers of fiber than in the past. If you think about it, everybody from the largest corporations to the largest government entities—we do a lot of work for the US government, the military—down to individual single consumers have access to fiber, either directly or by one step removed through a smartphone or a cell phone. That is a tremendous amount of distributed demand and opportunity to feed our networks. In addition to that, it creates a greater opportunity for us to build fiber more economically. We need anchor customers to build fiber, but if you have a lease-up strategy on top of it, you can make those anchored customer opportunities work, and you've got an exponential increase in the number of customers you can sell to. Those lease-up models work a heck of a lot better than they did back in the 2000s. They're a lot easier to make work today. I think that's how I wanted to say it. So I'm about to wrap up, but the point of a lot of that is the promise of the demand of the early 2000s is here today. All of those connected on-ramps, all of those customers, all those use cases are driving massive amounts of demand onto these long-haul networks. So the new Uniti: we're going to have a big 200,000 route mile network. I mentioned Joe Scaria. Is Greg Hotel in here? Greg's not here. Greg runs the wholesale business at Uniti. I see a bunch of Uniti salespeople and support people. Go find them today; they'll tell you all about that big, beautiful network. We're selling waves, we're selling dark fiber. Tremendous amount of opportunity. I think for other wholesale long-haul providers, their networks are going to benefit from a lot of these same trends and themes. So a much more rational and healthy market today. In addition to that, we're very excited at Uniti to have all of these distributed endpoints, all of these on-ramps. We're going to own ourselves 5 million connected endpoints through Kinetic and fiber to the home, through our enterprise business and connected buildings, through connected towers, connected small cells. Again, all of those endpoints driving traffic onto our wholesale network. Importantly for us, and like a lot of communications providers, our endpoints are in relatively unique positions. We're not really competing with overbuilders; we're not overbuilding others. We're targeting white space where fiber doesn't exist today. In our view, if you're first or early to build fiber into unique locations, you have a right to win many years into the future. So we're very excited about where we're positioned from a Uniti point of view, and super excited about where we're positioned from an industry point of view. I think the more we stay focused on rational business models and deploying capital in a smart way that actually generates returns, we're going to continue to do well and prosper. So Chip, I think that's all I had. I'm ready to take any questions.

Yeah, it's a great question. A lot in there. I probably will miss some of it, so keep me honest. Well, first of all, I think AI as an opportunity as a demand set is definitely real. We talk to the hyperscalers daily, weekly. They are believers in it. Those opportunities are finding their way into our funnel, into our bookings, into our business, onto our networks. You don't just flip a switch and turn on a network; you have to plan well in advance with your providers, with your vendors like us. Also, just listen to what they say publicly. They talk about needing to spend hundreds of billions of dollars to invest in infrastructure to enable AI and generative AI. And they also say—and this is the part I love the most—that even if AI doesn't materialize, they're going to use that infrastructure for other things. The data center capacity, the fiber capacity. They see these use cases driving demand and driving the need for infrastructure for themselves. So I think you're spot on to include them more in this Encompass dialogue because they're going to be critical customers of ours going forward, and they're going to drive where we make investment decisions going forward. So I'm a big believer in that as a customer set and that demand set. There's a lot of debate right now about the build-out of this learning phase and what that means for our industry. You've had companies like Lumen and others announce large deals that have gotten a lot of focus from the industry. There are some in our industry who talk about building to these single-use data centers, whether or not that's a good use of capital, whether that's a good thing to do for a fiber company. But I'll tell you, at Uniti, we're big believers in it. Anytime we build fiber, there's got to be a lease-up case on top of that, back to the point about having a rational industry. But if we're building to a data center that by the way is usually going to have anywhere from two to four redundant fiber builds into that data center, and that data center is in or near one of our core markets, we're going to get a second, third, fifth, tenth use of that network over time. So these hyperscalers can be terrific anchor customers for us to help us build new infrastructure. I think for the first time in a long time, we're going to be building a lot of long-haul fiber and connecting routes. We've announced over the past few weeks some new long-haul routes that we're building in the Southeast where the hyperscalers are the anchor customers. So I think you're going to continue to see more of that, and I think it's terrific. I think the inference phase of AI, however, is when we're really going to see the demand hit our networks. All of those distributed endpoints like fiber to the home, the towers are going to really need upgrading, the small cells. I think that's going to be one of the catalysts for small cells to really finally hit their stride because you're going to need densification in a lot of these metro markets. So the investment in the inference phase of AI is really going to drive a lot of demand, it's going to drive a lot of traffic onto our networks. But beyond that, the use cases—like I said, I'm not a technologist, I love to sit around and think about it and talk about it. And man, on the robots, I thought, I want a robot. I think that's one of the great things about a lot of these use cases: they're relatable to us. I know a smartphone is better than a cell phone, and so does my wife. I don't need to explain it. I thought, I want a robot. I'll sell my wife's car and get a robot. So I don't know what all the other use cases are, but as an infrastructure person, I know that I want our fiber network to be near where that demand is, and I want to be talking to the customers that are going to be making those decisions about use cases and planning with them about the demand they're going to need. Thirdly, I got to know that our network is robust and has the capacity to support that demand. We were selling six and 12 strands to hyperscalers a few years ago; now we're selling 864-count strands. That's a big difference. We're selling multiple conduits. So you got to know your network is upgradable, has the capacity, and the core—the core, the core, the core. That's a military joke. But the core also has to be capable of supporting all these use cases and upgradable. So again, long-winded answer. The last time we tried to predict the future in 2017, we were so... I'm going to stay away from stuff I don't know. I'm going to just focus on the infrastructure and just know that there's a lot of use cases out there that are going to drive demand.

Yeah, exciting topic. First of all, we built that fiber in Mississippi for those data centers. Thank you for that. We did not rehearse that. Uniti's model always has been to build mission-critical infrastructure with a focus on fiber in tier 2 and tier 3 markets. So we've always focused on the non-NFL cities. Our view is, like I said, if you build fiber first or early in those markets, you have an opportunity to win for many years into the future. It's really part of the strategic rationale of our combination with Windstream: getting into those tier 2 and tier 3 markets at Kinetic and building fiber to the home in less underserved markets. It's always been our strategy, it's always been a successful strategy. It's starting to become a strategy that's more in vogue because more and more people are looking at the white space in these more suburban and rural areas and seeing that opportunity where there's not a lot of fiber. Number one. Number two, I think there's been a demographic shift, back to the COVID comment. I think pre-COVID, during COVID, and continuing after COVID, there's been a move out of the major metros into the more urban and suburban areas. So again, back to the point about is there demand if you build the fiber? I think increasingly there is. I think that demographic shift has benefited. But thirdly, and I think really to your question, Chip, there are more and more anchor customers that allow us to build into these more rural areas. Fiber to the tower, for example. T-Mobile building into the more rural areas as a result of some of the commitments they made on their Sprint deal. The other carriers as well. So building those networks out into those regions has been something happening for years now. We have more fiber-to-the-home providers building into some of these more rural areas with government subsidy or without. But if they're doing that, they need backhaul, they need metro, they need to connect back to the core. How do they get there? They use networks like Uniti and others that are building in those areas. But those are anchor customers that are buying dark fiber, they're buying waves. Terrific customers. Thirdly, yeah, hyperscalers are driving investments outside of the major metros, like I was saying earlier. If a hyperscaler wants to build a single-use data center in the middle of Huntsville, we're going to build it, and we're going to build rings around it, and we're going to connect it back to Montgomery or Mobile, and it's going to be terrific for everybody. But they're also building outside of these metros, and we can build around it. When we collaborate with the power companies, they're seeing the same thing, and it's allowing them to expand their grids, expand their transmission capacity. Listen, when the hyperscalers are buying nuclear power—I mean, robots and nuclear power, that's kind of what we're talking about, folks. I think that suggests yes, there's a tremendous amount of demand that's going to be generated in those data centers that has to be supported, and it has to then be supported by communications infrastructure, including and especially fiber.

I think it's a great question. In telecom, in my experience in the past 20, 25 years, there's always been this push to get bigger and the benefits of scale. Scale matters: cost savings, cash flow savings, economies of scale, etc. But there's also been this sort of balancing effect: if you get too big, you lose focus on the market and the local market. So as an industry, we've gotten bigger, we've gotten smaller over time. Right now, what I think is really exciting about the trends I'm seeing, at least, is that customers and what they want are driving M&A. It's a much more rational market, including wholesale. I think our customers want big, beautiful networks. They want to work with providers who have big scale, expansive networks. That's driving a lot of the consolidation that we showed on that slide earlier about all the logos back in the early 2000s versus a fewer number today. So I think that trend has been driven by customers. On convergence, I think it's the same thing. When you listen to the wireless carriers, they are believers in convergence because they're seeing it as a demand from their customers. Their customers want a fiber home connection and a wireless connection. I believe that, again, as a consumer, I can relate to that. So I think that is driving the consolidation you're seeing: Verizon acquiring Frontier when they sold Frontier 10 years ago. It made sense for them to sell it 10 years ago; it makes sense for them to buy it today for that very reason. So from a consolidation perspective and from an M&A perspective, I think that's going to continue. I think we're probably going to wind up seeing a handful of true fiber-to-the-home owners, similar to what we see in cable. You've got four or five large cable companies that have kind of divided the country. I think you'll see the same thing on fiber to the home. In the meantime, what does that mean for companies like Uniti and others that own those businesses? Number one, it means you better invest in your network. If you don't and you're just sitting there with a bunch of copper, you're going to get your butt overbuilt, because there's a lot of capital that's chasing that white space. I think in a good way. So we're very focused on building fiber in the Kinetic footprint to overbuild that copper with fiber, to protect that footprint and give ourselves a right to win for many years in the future. Secondly, it means we're going to have to be plugged into the M&A market and the consolidation phase going forward, because I do think it's going to happen. Head in the sand doesn't work. You got to be out engaged with the industry, and that's exactly what we plan to do.

Yeah, I think when I say I'm not a technologist, I'm a student of technology. I'm definitely a student of the trends. We have to be as infrastructure providers. So I want to draw that distinction, but I'm certainly not Elon Musk, and I'm not the Google CEO. I don't know is the short answer. But what I do know is that people are going to want to continue to consume mobile broadband, and they're going to want fixed-line broadband. They're going to want both. I don't think mobile broadband is ever, in our lifetime, going to supplant fixed-line broadband. And to that point, there's a distinction between mobile broadband and fixed wireless. I think fixed wireless is a great product today, but it is not a product that's going to persist long into the future because it has limitations: limitations on latency, on demand. The carriers have only so much spectrum to use. That's why you see the wireless carriers investing heavily in fiber to the home. They know that those 20, 30, whatever 40 million fixed wireless customers are going to be looking for a home in three, four, or five years, and they're going to be looking at cable or fiber to the home. I think it's going to be fiber to the home where those customers are going to go. So as an industry and as infrastructure folks, I think we got to be ready for that. To me, being ready for that is having these big, robust wholesale networks that can support that demand, whether it's coming from mobile broadband or fixed broadband. Get ready for it. And two, pick your spots, invest in these white space areas that don't have fiber today, and you're going to wind up benefiting, whether it be as an operator or eventually part of some consolidation phase.

Well, I'd love to hear your answer to that question, Scott, because I have a lot of respect for Scott and have known him for many years. I'm going to listen to your panel; I can't wait to hear it. I think it's a good thing in some ways and a bad thing in other ways. I generally believe in less government involvement in all things, including especially our industry. I think government subsidies to get more fiber out to more rural areas, in a vacuum, is a good thing. That's a smart thing. In many ways, it's another anchor customer. You can talk about the wireless carriers or fiber-to-the-home providers or hyperscalers as anchors. If you're using government subsidy, that could be another anchor customer to get out into some of these more rural areas. I think the rubber meets the road, however, with what are the red tape associated with that subsidy, what are all the conditions associated with it. That's why we love Chip and Encompass, because they help us as an industry interact with policymakers to help them understand the implications of this investment. But I think on balance, it's a good thing, but it has to be used in the right way. It can't disincentivize existing providers, it can't disincentivize incumbents, it can't be used to overbuild where it doesn't make sense. Let's use the dollars in parts of the country or parts of the ecosystem that actually need that investment. When you look back at some of the subsidy programs that have happened over the years, a lot of companies have gotten subsidies, spent a lot of money—our taxpayer money and theirs—and built networks that didn't really make sense. A lot of those companies aren't here today. I hope and think that we're learning from a lot of that historical investment, and in this BEAD rollout, we're going to have much more successful investment. But we got to stay engaged with Chip and Encompass and others to make sure that happens.

No, it's definitely on our radar. In fact, a few years ago, we had a construction business within Uniti that was specifically focused on building fiber for Departments of Transportation around our footprint. At the time, it was single-use investment, so we'd build it and turn the whole thing over to you because that's what the Departments of Transportation require. But we sort of thought, okay, let's do this, let's get close to the various DOTs, and eventually we'll turn this into building for all those autonomous vehicle fueling stations that you mentioned. That didn't work out for a variety of reasons, but we were thinking about it and continue to think about it. I absolutely think it's one of those use cases. When I see what autonomous vehicles are today, even in my own vehicles, I can drive for long periods of time with no hands on the wheel. You sort of relate to the value of that, and you can see it coming. In order for that to happen, you're going to have to have tremendous amounts of fiber. I think the challenge is going to be that you're going to get into government regulation, dealing with the various Departments of Transportation, dealing with the federal government on interstates. I don't know the solution to that right now. That's going to have to get figured out, and it's going to be a necessary precursor to autonomous vehicles really ramping up.