Episode 23: Anthropic’s IPO, T-Mobile’s Physical AI and Verizon’s FWA Push

Peter Jarich (00:00)
Hi everyone, and welcome to Intelligence & Noise, the podcast from GSMA Intelligence, where we shift through the week’s news. We try to pull some intelligence out by looking at everything that might seem like noise, but maybe it’s not. We do it through the lens of the CEO newsletter that comes out from the GSMA. I say every week, if you get it, congrats to you, you’re important. If not, don’t worry, that’s what we’re here for. We’re here to tell you what’s in the newsletter, what could have been in the newsletter but wasn’t, and why it all matters.
So I’m Peter Jarich. I head up the GSMA Intelligence team. I have with me Tim, who heads up... just does tons of stuff here. I don’t even know what. Reviews are coming up, so we should probably figure out what exactly your title is. Tim does everything.

Tim Hatt (00:44)
I do something. I play hockey and I kind of do something useful.

Peter Jarich (00:54)
Tim is the designated “Peter gets hit by a bus” person, and we don’t even know what happened.

Tim Hatt (01:04)
I think we call that a utility infielder in baseball, Peter.

Peter Jarich (01:07)
Utility. That’s very Canadian of you, that’s really nice. So we’re going to dig into it. It’s been a big week. When this hits, it will have been a big week for a while. We’re going to dig into a few things, but as often, I kick off with a bit of a story. Tim, it was Halloween. By the time this runs it’ll have been Halloween for a while, but it was just Halloween this past weekend. Did you have trick-or-treaters come to your house?

Tim Hatt (01:32)
I did. At this point I am a chaperone who gets told to just keep my distance and not cramp styles, which is what I did, but I always enjoy it.

Peter Jarich (01:42)
We understand Halloween is picking up steam here in the UK, and I understand it’s picking up steam in the US. I listened to a podcast that said spending on Halloween decorations is up almost ten, fifteen percent year on year, which, given the cost of everything, may just be inflation. I don’t know. But it’s interesting. We don’t have kids, but we had folks come over to get candy. We just do it out of the garage because we don’t want to keep going back and forth to the door.
It’s interesting because, as always, and maybe you saw this too, Tim, it feels like there’s a good distribution of kids. There are kids who take it really seriously, and kids who are just going through the motions. They throw something on so they can go and grab some candy, or their parents just want to take them around to get candy. And there’s no shame in that. But I’m struck every year by this mix of the really dedicated ones and those who are doing it because they want the candy, and this is a step to get there. They maybe don’t even bother with a costume, or pretend to do a costume.
That’s important because, as part of the first story, or maybe a second part of the first story, I think there’s something going on in the industry that is probably a very big going-through-the-motions exercise. Someone has made a filing just because they need to make a filing, whether or not it’s actually necessary to convince people to invest in them. So it might be a hundred-billion-dollar going-through-the-motions exercise.
But there were two stories this week in the CEO newsletter. One was about telcos getting into healthcare. We’re not going to talk about this much, Tim, but it’s kind of important. It was about DT investing in Tandem Health, which builds AI solutions for the European health tech industry, AT&T doing smart meters for remote monitoring via eSIM, and was it Telefónica picking up 10% of OGA, a company that’s delivering AI into various verticals including health tech? We’re not going to talk about this much, but what’s your thought on why it was even worthy of making it into the newsletter this week?

Tim Hatt (04:15)
Well, I think it’s worthy because the healthcare industry is a big industry, and it’s an inefficient industry. It still accounts for a huge amount of government spend, often over ten percent of GDP, even in parts of Europe. And when you look at our enterprise survey data, they’re going to spend about 10% of their revenues on digital transformation over the next ten years. How do you make things more efficient? Telehealth, remote monitoring. And I think these stories, with eSIM especially, and the AT&T one, speak to that promise. It is complex, but it’s moving.

Peter Jarich (04:47)
Yeah, and what’s interesting is the other part I would call out. They’re also possibly a little behind in things like AI, which makes sense. It’s a regulated industry, and you don’t want to just play around with new technologies, so they’re naturally going to be a little slow on some of those things. But that means that as those technologies mature, and they know they need to take advantage of them to be efficient themselves, there’s going to be an opportunity there, if telcos want to play to that.
It’s also nice because it shows that, as much as we looked at 5G as an opportunity, build it and they’ll come, with verticals rushing to us, you have to make those investments. It’s not just going to happen on its own. It happens via investing in these companies and getting closer to them. So it was good. At its best, the CEO newsletter does a good job when it calls out things folks might not have seen.
But it’s worth going into the news that everyone probably did see. The big headline news was really around telcos getting into AI and driving agentic solutions, and I think it did a good job by highlighting a couple of different stories. One, Verizon calling out a deal with AWS as part of their earnings call, where they talked about a $2 billion deal for dark fiber that adds to work they’re already doing with Google. They suggested this is not the end of it: there’s more hyperscaler and more neocloud pipeline. The second story was T-Systems talking about supporting Banco Sabadell in Spain with their agentic AI platform.
But the big one, and you called this out even before the newsletter, earlier in the week when we were sharing it on email, was a big old... I don’t even know if we call it a blog. There was a post from T-Mobile US about physical AI. I feel like we’ve got to have some definition: five hundred to seven hundred words is a blog, seven hundred to whatever is something else. But this was a lot. They had a lot to get off their chest. You sent it to me and to some other folks internally. What struck you about this as something to say, “Hey, pay attention to this”?

Tim Hatt (07:32)
Well, it’s actually not even one story, it’s multiple stories within a story, I think is the first thing. T-Mobile in the US has signed a bevy of partnerships to further advance its agentic and edge inference strategy. They’re working with a group called Figure AI to help orchestrate workloads from the device to the network. They’re also working with other partners to help manage the development of humanoids and how their network copes.
And Peter, why this is interesting is that it’s two points, and I’d like your reaction to them. One is that agentic is really the next outpost on the spectrum of where operators move. We started with early activity on network efficiencies, then tested the viability of the edge, and now we’re pushing out further to the device. This is an example of that. Two, it shows what role an operator can play. That’s always been a little unclear in technological shifts. Operators are bigged up as having a role, but it’s not quite clear if that manifests. Here you’re starting to see a central role in the value chain, and having the right partnerships to orchestrate that stack.
So when I look at that set of T-Mobile partnerships, and we’ll get on to the big news on agentic, really we’re looking at the use of a 5G Advanced network, deterministic connectivity, and having AI orchestrate what is in reality a very complex set of workloads. And I’ll be honest, I was in touch with the senior leadership on the technology side of T-Mobile last year, at the start of their edge journey. They were mapping this stuff out, and it was clear at that point that they were thinking big picture. And when you see this blog now, it sort of comes together.

Peter Jarich (09:32)
Yeah, and it’s a bit of recapping a lot of things that folks have said before. Physical AI is the next big thing. We’ve talked about generative, we’ve talked about agentic, and now physical, and I get it, we’re always looking for the next thing. So, moving past agentic, what does it look like with physical AI? There’s an opportunity here because you can’t do everything on the device. You need the network, and you’ve got to have the right workloads in the right places.
For me, a couple of things stood out as particularly important. One, they did a good job of linking this to real work they’re doing. I’ve said this for years and years, and every analyst says it: we love announcements that talk about what people are actually doing, much better than when you don’t have a customer. So, the work they’re talking about doing in San Jose, with Linker Vision and Laplace Intelligence, looking at connected cameras. We’ve talked about the connected camera, edge AI application multiple times in the past, so it’s a well-known application. San Jose has been big as one of the US cities driving IT modernization, so they’re doing some innovative things. But that’s important because it shows it’s not just robots. It’s not just cars and robots that we’re all talking about. And then they make a nod at the end to 6G and the network as part of the AI infrastructure, which I think is important. So it’s good because they’re talking about robots, but also sensors, cameras and all of those.
There are a couple of things they could have done better. There’s a bit of a reference to uplink and low latency, but I would have expected a bigger play on the uplink side, particularly given what they’re doing with their network. They have been so strong on talking about the capabilities of their network. It’s been such a big part of their messaging. I was also surprised APIs weren’t really part of the story. And there’s no real mention of reliability, trust and safety, or sovereignty, and how that fits in. Obviously the sovereign story means something very different in the US versus in other markets, so it’s not going to be such a big deal, because it is the US. But think about this in some other market, where I’m relying on things like cameras that are very important, maybe mission-critical, maybe for public safety, and we’re doing orchestration that is going to live on a public cloud that I don’t control. Maybe it works in the US. Maybe it doesn’t. And I’m not saying this just because T-Mobile US is connected to DT, and you definitely see that sovereignty message from DT.
As we refine this message around physical AI, given that we’re talking about physical things, and it’s not like you can’t do a lot of damage in the digital world, but when we extend AI to things in the physical world, that safety, security and reliability part becomes even more important. So as we tweak that message, I think it’ll be important to see that. But it was good, definitely good, thought-provoking, and definitely part of AI week last week.

Tim Hatt (13:24)
Yeah.

Peter Jarich (13:25)
And it probably was AI week, because of the big news that I guess we want to link this all to, which is probably the bigger piece. Towards the end of last week we saw a whole bunch of different fundraising. Meta was raising about a hundred billion, ninety-seven point five, with an equity raise to fund some of its AI development. Anthropic was doing 63 billion in infrastructure financing for Broadcom infrastructure, including some senior debt, some of it backed by Broadcom itself. Go figure, that seems to happen with companies funding their own purchases. There were smaller things in other markets too, with Nebius raising 10 billion from Brookfield. A whole bunch of raises, all at once. And I guess it kind of made sense, because the big news from last week in AI world was what? Tim, what was the thing that even your friends and neighbours were probably talking about?

Tim Hatt (14:39)
Well, it is the Anthropic IPO announcement, which came out last week. You’d have to be living under a rock in the tech world to have missed that. It came out slightly delayed, but no less of a bang than what we expected.

Peter Jarich (14:58)
Yeah, they filed confidentially in June, so we’ve known about this for some time. It’s interesting. In all my years, the amount we’re talking about S-1s this year, where we finally get a view of what’s going on with the curtains pulled, is striking, and it’s interesting just to pore over all of these. But yes, filed on October 26th. I know you and I have each gone through our own fair share of facts and our own 318 pages, 97 of which are focused on risks. So not quite a third, but a good thirty percent focused on the risks. The symbol will be “ANT”, which is a choice, I guess, of some.
What I think is interesting is that, given the filing timing, they can’t start the roadshow until the 10th of November, but there’s an expectation that trading could start in the second half of November, which means within a month. That seems pretty quick. It’s going to remain a public benefit corporation. A Founder LLC will be set up which will hold 50.1 percent of the voting shares, with the seven founders, so the folks at the top. There’s been no discussion of the number of shares, and it hasn’t been disclosed how many shares there will be with the pricing. There seems to be an expectation it’ll be around a hundred billion raised.
We all have lots of thoughts, Tim, and so much has been said that at this point there’s almost not too much we can say. We will try to bring this back to the telco world, because I’ve got some thoughts there. But when you heard this, or had a chance to pore through it, what struck you?

Tim Hatt (17:00)
Well, number one, it’s huge. You pointed to that. At a hundred billion raise, that’s a valuation between one and a half and two trillion, with a T. That would be the biggest in history. If you go back to their private valuation back in May, it’s up like 2x now, so you can see how quickly things are moving. So the scale is one.
Two, whether we’re talking about Anthropic or OpenAI or another of the bellwethers, it gives us a window into the AI world in general. You look back at some of the stats they put in, and it’s interesting, for better or for worse. You look at the revenue growth: H1 is up over a thousand percent in the last year. You look at the churn, and at how much money flows through the cloud hyperscalers right now, a lot of it. Then you start to see, hold on, there’s a lot of dependency built in here in the AI world. It’s not just one company. What you’re really betting on is an ecosystem.
And thirdly, as you called out, the founders maintain control here. Just like with SpaceX, where you’re betting on Musk, here you’re betting on the Amodei family and their friends to run this through. And whether you like that or not, that’s what’s on the table.

Peter Jarich (18:18)
Yeah, to your point, thirteen hundred percent year-on-year growth in H1 revenue is almost meaningless, because it’s going to change every quarter. It’s almost more important to see what Q3 looks like, but that’s also going to be meaningless because we’re still on such a weird trajectory here. Compute costs are falling: they used to be two hundred percent of revenue and are down to fifty-five. Obviously you want to get the costs lower than the revenue. They have almost 12,000 customers, with 250 spending 10 million plus.
Two things were particularly important for me. We’ll get to the ownership side of things, but net revenue retention was 900%, which means customers are coming back and spending a lot more. It’s not GSMA Intelligence levels, let’s be fair.

Tim Hatt (19:13)
You’ve got a ways to go to get to that.

Peter Jarich (19:15)
But still, six point six million per customer is, if you look at the numbers they talked about earlier, just up incredibly. Again, not quite GSMA Intelligence levels, but they’ll get there. They’re still a young company. What I think is interesting is that they call out that, at least for their Fortune 500 customers, they estimated that Anthropic represents less than 0.1 percent of expenses. It’s an interesting way of saying either there’s room for growth, or there’s room for growth in other companies. It’s an interesting way to ask how sustainable this is and what the size of these numbers means.
I’ll go back quickly to my first comment. My understanding is that the point of an S-1 is to give prospective buyers an idea of what a company looks like and an understanding of what they’re getting into. That’s why you spend so much time on risks. When I talk about this being sort of going through the motions, I don’t think the risks are unimportant at all. But it’s not like any of those prospective buyers are looking at this and going, “Huh, should I put money in this?” I’m pretty sure everyone is lined up, and that’s why you can see potential trading from the second half of November. I think those risks are interesting, and I give them credit for calling them out in such detail. I think that’s probably one of the reasons Anthropic has such goodwill: the whole thing has been about being responsible. Yes, they’ve had some issues, and people will question who they work with and who their customers are, whether they work with the military or not. But I think they were seen as the responsible ones, as trying to be good with this. So I appreciate all those risks. But it’s not like they have to file this in order to go public and are just telling us things; it’s not like this is helping anyone decide whether or not to put money down.

Tim Hatt (21:40)
Yeah, exactly that. And I want to come back to something you said. You talked about some of the risks they highlighted, and it’s stark when you look at the level of transparency and candidness they are giving to the bad side of AI: the risks, the worry that this could end up badly for humanity. We need to get that out there because we’re asking for a lot of money.
I’ll be a little bit cynical here, Peter. I’m an optimistic person, but I think part of that is legal advice on getting that warning out, for the incidents of security or other kinds that are inevitable. Not just possible, they are inevitable at some point, so you’ve got to get that out. But in doing so, the deeper concern is genuinely whether there are existential risks here, and not necessarily the end of the world, but serious risks to the economy, or through security, or to governments, that are just going to have to be accepted and mitigated through this process. I haven’t yet seen a mechanism for that, so that will be an interesting part of the IPO and of where they go in year one. How do you feel about it?

Peter Jarich (22:55)
Yeah, I think what’s interesting is the risks, and you’re right, a lot of those risks are just about making sure they’re calling things out. What’s interesting is that it runs the gamut. From “we may not have enough compute capacity out there, it’s a race to get compute capacity, and if it’s not there it may stifle our growth,” to questions about regulation, to questions about capital intensity, and then, to your point, to “there may be existential risks to humans.”
And for the record, for everyone at the GSMA, when I go to the investment committee and ask for money, look at my 2027 business case. I listed out the risks. “GSMA Intelligence bringing down humanity”: not one of them. So just remember, folks, that’s a plus for the business, Tim. None of the risks I called out for the 2027 business case include potentially creating the end of humanity. So, point for us.

Tim Hatt (24:13)
And with SpaceX, with Anthropic, with Meta, with a huge amount of what is the American economy and the global economy, power is concentrated in a very small number of hands. So we’re working with a huge amount...

Peter Jarich (24:28)
I agree. And I think my big concern is not about the ownership. So, yes, fifty point one percent compares to sixty percent for Meta with Zuckerberg, and eighty-five percent with Musk. And I do feel, in part because of the way they’ve operated, that this is part of an “okay, now that we’re public, if we decide as a company that we need to slow down development, or do some things because they’re the right thing to do, those folks can make that call,” as opposed to a bunch of shareholders going, “No, no, no, keep pushing hard. Yeah, I saw that thing you listed saying this could bring down society, but just keep going.” So I do think there’s a bit of that, which I guess means I’m also being optimistic.
My bigger concern is the sustainability side of things. Clearly that nine hundred percent improvement in retention from customers is crazy. But the bigger one is that, as models improve across the board, their huge benefit and huge differentiator has been the development of crazy frontier models, always at the front, developing new capabilities that are important for folks. What I don’t know is, as open weight models become more important, when is good enough good enough?
I’ll come back to that, but I think we’re going to get to a point where, as folks look at their budgets, and they have started looking at their budgets, they recognize that they need to use different models, and some models may be good enough for what they need to do. Then I don’t know how much of a differentiator it is, or how important it is, to have those frontier models. What I do think is interesting is that you end up with a potential flywheel. You get more capable AI, which drives customer demand. That drives compute investment, which means more capital commitments. And then you get more capable AI, which drives customer demand again, and you get this great flywheel. That’s the hope.
But there’s this other thing that’ll be interesting: capable AI leads to societal worries and misuse like we’ve already seen, which means regulation, which means potentially slower releases. If that’s the case, then those good enough models become good enough and that differentiator goes away. So I think we need to see how it plays out.
It’s important, though, that we talk about the telco angle here. For me there are two. One, I’ve talked to a number of folks over the past year who bring up that exact question: for what we’re doing, is good enough good enough? Do I need frontier models? We’ve seen this from the filings, and from operators who say the huge benefits, the biggest gains, are in efficiencies. Do I need the best of the best frontier models to do customer care? And you’ve seen what AT&T is doing with the OTel models that they’ve delivered, open weight. I think that’s a good example, and a good use case of good enough being good enough.
The second thing, from the telco side, is that so often we hear people looking at the telco world and going, “You guys are slow, you’re there on the connectivity side, but you’ve really lost the thread. Look at the way the big tech guys are growing.” I think when you read risk factors which include “this may cause huge societal harms and may bring the end of humanity,” telcos are also safe, right? These are companies people’s pension plans are invested in, national companies. And this is a reminder of why, on some of these cutting-edge things, you’re always going to have that constraint from a telco: when those risks get that big, that’s got to be a place where you step back and say, “Okay, maybe I watch, and I figure out how I can benefit from these.” I just don’t see...

Tim Hatt (28:50)
I think that’s well said, and I would just add two others from a telco angle, real quick, before we move on. One is that telcos will want competition in the AI space. They will want to have a degree of choice. You’re right to say there are telco-specific LLMs, and those are being very successful. They’ll also want competition, and I think a number of verticals will push for that, on the security side and the sovereignty side.
We saw the Mythos incident earlier this year, and we’ve seen other Anthropic-related concerns emerge. To their credit, Anthropic are being candid about that in the IPO, though it’s a different thing in practice. And to your point about telcos being critical national infrastructure, making sure that AI security is absolutely paramount.

Peter Jarich (29:38)
And that is front and center in people’s minds, right?

Tim Hatt (29:42)
It is. Yeah.

Peter Jarich (29:43)
It is a thing. It’s stuff that folks are thinking about. Honestly, just as a brief anecdote: I was listening to a podcast this weekend. There was a tiny snippet that I wanted to remember and couldn’t. I was on the couch on a Saturday, and I wanted to pull it up for my wife. I went to the personal AI agent I use and said, “Is there a transcript for this podcast?” Yep, found it. “Can you find this reference?” Yep, here it is. “Can you clip the audio for me?” And it said, “Yep, I’m going to go to Hugging Face. I’m going to connect with Hugging Face to do this.” I mentioned it to my wife, and she got scared: the mention of AI going off and doing something with Hugging Face was enough, in the back of her mind. She was like, “You sh...” I said, “No, I’m sure it’s fine. It’s not hacking Hugging Face to find the right capabilities to...” In the end, it clipped the audio for me, which is great. No idea how many trees burnt down to do that, but I think it is in the public consciousness, so that when you say these things, it’s top of mind.
Two stories were not chosen. One: the Swedes are beginning to build some capabilities and test beds around 6G and advanced innovation. I think that was announced earlier in the year, but they’ve finally started launching it. The other one we want to touch on briefly is Verizon. Verizon, in their Q3 earnings, talked about basically a new FWA service using millimeter wave to deliver one-gigabit service. I think the way we would have referred to it in the newsletter, if it had been included, was “taking the fight to fiber.” I can see why it wasn’t included against the backdrop of everything AI and healthcare, but it still feels like we don’t talk about FWA enough, do we?

Tim Hatt (31:50)
We don’t, and we should do more, because FWA is an unsung hero in the 5G toolkit. I think we can talk about 5G monetization as being mixed, but FWA has been an unsung winner. And Verizon here is doubling down. They’re looking to expand the FWA service on millimeter wave into US cities. Ten to fifteen million premises is the goal, on top of the existing target of forty to fifty million.
And when you look at our numbers, we think FWA is about fifteen percent of broadband share in the US by the end of the year. So this is not small beer, it’s already eating into cable share. And for an operator, it’s an economically efficient way of providing service because you’ve already got the infrastructure in place.

Peter Jarich (32:38)
Yeah, we did some work, a few years ago now, back when some of the initial FWA solutions were being rolled out. I’d recommend folks go and look at it. It’s a TCO analysis that is not hand-wavy, not back-of-the-envelope, looking at urban, suburban and rural, Europe, US and Latin America, with different models to understand what it looks like. The downside is you end up with some pretty complicated charts, looking at three different markets, three different types of area by population density, and comparing fiber to the home, whether ducted fiber or owned fiber, versus FWA with indoor, high-powered or standard CPE. So lots of different comparisons. There’s some great chart work where you’re trying to show 17 things at once, not quite that bad. But the nice thing is there’s one line which makes it pretty clear: below it means you’re seeing a better return from FWA, and above it means a better return from fiber. And in most of the cases, you see a better comparative return from the FWA side.
Again, this brings out that there are a number of different plays. Are we talking about indoor CPE, outdoor CPE, high-powered CPE? Your mileage may vary, and every market is going to be different, but it does highlight that with FWA, with millimeter wave, and I think the other interesting piece called out by Verizon is that by using this, it also frees up C-band capacity. Taking some of the FWA they already have off that spectrum allows them to use that spectrum more efficiently. Win-win.
Tim, we’ve got some travel coming up. This is the last segment, the Willie Nelson segment. There’s not much on our minds, since the Anthropic stuff has taken up every brain cell I have. So the “on our minds” part is gone, but on the road: I just got back from Huawei. Look for some notes on that soon. We always get a good view of what’s going on in China and the progress being made there, and it’s always interesting to see how capabilities are being driven. It’s a bit of a different message from some of their competitors, many of whom are talking about AI. You’re seeing a lot of focus on materials science, and on process engineering and manufacturing to get different efficiencies. So while the rest of the world is looking at software and digital and trying to figure out what they can do to improve capabilities, you see a lot of process, including leveraging diamonds and sapphires for cooling, which got a lot of discussion.
I think a lot of analysts even started talking about whether they are lab grown or not. Of course they’re lab grown, which is interesting. And since you talked about being cynical, the cynic in me found an interesting article that popped up later in the week saying China supplies about half the global lab-grown diamond supply. So is Huawei using diamonds because they’re the best, or because there’s a national agenda to say that gross margins are pretty good on industrial diamonds and maybe you should use some? I don’t know, but it does sound cool.
You’re off, Tim. So again, look for some insights from us on that. It was a good event. We had a whole bunch of the team there, about six of us. But Tim, you are headed off to Doha soon, correct?

Tim Hatt (36:46)
I am. I am off to Qatar with my colleague Pablo Iacopino and we’re excited. We’ve been to Doha three years running now. We’re running a number of events. We’ve got roundtables, there’s stuff on 6G, stuff on enterprise, stuff on satellite. For me, Peter, what I’m looking for from the event is, number one, a pulse check on industry health. I think the Middle East is interesting in terms of how it’s pushing the 5G agenda, so it’s an interesting sense check. Two, the satellite and sovereignty side is going to be key. We’re seeing a lot there, and we’re going to be hosting a summit on that, which we’ll watch closely. And then, three, the general outlook, and how the Middle East looks geopolitically in terms of East versus West. Right now they kind of straddle both: a lot of trade with China, a lot with Europe. How does that dynamic manage in the current...

Peter Jarich (37:48)
It seems a bit trite, but I think going to these places really helps to bring that home. I came back from Huawei in China through Doha, and even just going through the airport you can feel it: you see a much bigger diversity of folks going all over the world than in many places. Obviously that’s going to be reflected in how they do business.
Right before you go to Doha, I am off to the US, where I’ll be at the Brooklyn 6G Summit, as well as FutureNet. This is the first time I’ve been to the Brooklyn 6G Summit. I’m going in part because early in the week I’m headed down to our Atlanta office, where our managed services team is doing a user event, so I’ll be speaking to those folks. They do great work and have a lot of great products, so I’m happy to be there. Then I’ll head up to New York. The Brooklyn 6G Summit, sponsored by Nokia, is going to be a lot of AI, AI, AI, not surprisingly, but it’ll be good to see how the non-technical, maybe business, side is being discussed. And then, of course, FutureNet always does a good job talking about what’s going on in the core side of the network and in monetization. So it should be a whirlwind, which means that by the time we get back to the next one, we’ll have a lot to share. Thank you, Tim.

Tim Hatt (39:14)
Yeah.

Peter Jarich (39:15)
Thank you as always, everyone, for tuning in. Please take some time to send us notes and comments, and tell us what you like. More importantly, take some time to rate this. Like it, don’t like it, take some time. If you’ve listened more than twice, please rate it. That helps people actually get to us. Tim and I aren’t just talking for our own benefit. So please take some time to do that. Tune in next week. And in the meantime, I’d like to remind everyone, in the search for intelligence, don’t forget to embrace a little noise. Thanks, everyone.

Tim Hatt (39:53)
Thanks, guys.