NVIDIA: What’s Next, Beyond Market Maker?

IT Infrastructure Principal Analyst Angela Lambert and Senior Analyst and NVIDIA research lead Ben Carbonneau join “TBR Talks” to share key takeaways and insights into NVIDIA’s strategies for 2026 and beyond. Recently, “TBR Talks” host Patrick Heffernan met with executives and others within the GPU ecosystem at NVIDIA GTC, and on this episode, all three analysts share their thoughts on partnerships, business strategy and growth for the leading GPU firm.

Episode highlights:

• NVIDIA’s partnership announcements and partnership strategy

• NVIDIA chip manufacturing in the U.S.

• The competitive landscape for NVIDIA

• Ambition, risks and partnerships: Where will NVIDIA be a year from now?

Listen and learn with TBR Talks!

Submit your Key Intelligence Questions for Patrick and his guests

Connect with Patrick on LinkedIn

Learn more about TBR at ⁠⁠⁠⁠⁠https://tbri.com/⁠⁠⁠⁠

TBR Talks is produced by Technology Business Research, Inc.

Edited by Haley Demers

Music by Burty Sounds via Pixabay

Art by Amanda Hamilton Sy

NVIDIA: What’s Next, Beyond Market Maker?

TBR Talks Host Patrick Heffernan: Welcome to TBR Talks: Decoding Strategies and Ecosystems of the Globe’s Top Tech Firms. Where we talk business model disruption in the broad technology ecosystem, from management consultancies to systems integrators, hyperscalers to independent software vendors, telecom operators to network and infrastructure vendors, and chip manufacturers to value-added resellers. We’ll be answering some of the key intelligence questions we’ve heard from executives and business unit leaders among the leading professional IT services and telecom vendors. 

I’m Patrick Heffernan, Principal Analyst, and today we’ll be talking about NVIDIA GTC with Angela Lambert, Principal Analyst for TBR’s IT Infrastructure Practice, and Ben Carbonneau, Senior Analyst for TBR’s Infrastructure Practice. 

NVIDIA’s partnership announcements and partnership strategy

Angela and Ben, welcome back to TBR Talks. Through the magic of recording, it feels like we’re doing these on back-to-back days, but you won’t get to listen to them on back-to-back days, but that’s the way it goes. So last week, actually last week, I was down in D.C.- kind of a back home for me, having lived there for almost a decade, a long time ago. It was nice to go back, but this was for NVIDIA’s huge annual global, I think it’s GPU technology conference is what it technically stands for, NVIDIA GTC. And so got to see the keynote, hear a bunch of presentations, meet with a lot of folks, spend a lot of time with NVIDIA the same week that they catapulted over $5 trillion in their evaluation. So, I’m not saying I was responsible for that, but, you know, it was nice to be there at the time when it happened. So, what I wanted to do is bounce some of the ideas that I heard last week from NVIDIA off of you two, because this is a company that you both follow very closely. So, what I heard, what you think about what I heard, then maybe where we think NVIDIA is going next. So, for both of you, and I guess I’ll start with Ben on this one, the partnerships that were announced last week included Nokia, Palantir, CrowdStrike, and a number of other ones. Did any of them surprise you? Are there any departures from what you’ve seen from NVIDIA before?

Ben Carbonneau, TBR Senior Analyst: I’d say at a high level, nothing really surprises me from NVIDIA on the partnering front. I’d almost think more of what large technology companies aren’t partnered with NVIDIA. So, I see NVIDIA really being kind of, you know, at the epicenter of AI, right? So not only does, I think, like every large technology company want to partner with NVIDIA, but NVIDIA’s go to market strategy and being kind of that partner led, wanting to be recognized as more of a platform provider than a solutions provider, I think through that lens, they’re really dependent on the partnerships for go-to-market. And whether that’s partners to gain some sort of technological solution or partners for go-to-market specifically, or even partners around integration, I think there’s a lot of different flavors of NVIDIA partnerships. And I guess nothing really surprises me there. Nokia, Palantir, CrowdStrike. I think Palantir was a good one, given the venue.

Patrick: Of course, right, with the U.S. government, absolutely, yeah. How about you, Angela?

Angela Lambert, TBR Principal Analyst: One- looking at the partnership announcements, I have to wonder in relation to that $5 trillion valuation you’re talking about is if we’re going to see more of an increasing trend in some of the heavy investments that NVIDIA is making alongside partnerships. So as part of that Nokia announcement, NVIDIA is investing a billion dollars in their company to help develop this technology. And only a handful of weeks ago were they doing the same thing with Intel, investing $5 billion in the form of, you know, joint partnership, but also getting some stake in these companies. So, I think that to me, it’s indicative of the fact that NVIDIA has the capital to invest. And so, beyond your standard tech partnership where two companies are working together, they’re going to also be willing to invest significant amounts of dollars to help move the market in the direction they want to see it going.

Patrick: And part of that is NVIDIA investing for those companies like a Nokia to turn around and purchase chips from NVIDIA, right?

Angela: Absolutely, yeah. So, it’s mutually beneficial in the sense that they’re getting joint investment in developing technologies, but then on the backside, NVIDIA is guaranteeing purchases down the road of their technology.

Patrick: Right. And Ben, to your point about the number of partners walking the floor of the conference, it was pretty amazing. CDW was there, which I did not expect. There were a number of- Booz Allen Hamilton, a number of the other federally focused companies, which I did expect. Only one out of the big four firms were there. And I’m not going to name names. They can figure it out for themselves, and I’ve already been in touch with all four of the firms to say, well, three of them to say, why weren’t you there? And one of them to follow up. So, it’s curious who did show up and who didn’t show up for this event. 

Does NVIDIA run the risk of spreading themselves to thin with their partnerships?

But that sort of does lead to a bigger question. And in talking with some of the folks I met with from NVIDIA, they repeatedly talked about open system, providing developers tools, all about their partnering. Has that, and to me, it was just surprising because we talk about ecosystems all the time, but it’s an evolving thing. They talked about it as something they’ve been doing for a long time. Is it true? Is that consistent from them? And then is there a risk of them sort of spreading themselves too thin, diluting their value, making missteps if they partner with so many companies? Angela.

Angela: So, I think that this is a consistent theme from NVIDIA, and I can’t wait to hear if Ben feels the same. When I think about the potential risk of NVIDIA spreading themselves too thin on their partnerships, you know, we see them moving in so many directions, right? Beyond just the data center, you see huge investments in automotive, and telecom, and robotics. You know, the list goes on and on. I think to me, NVIDIA, rightly so, sees themselves as a market maker today. And frankly, not every endeavor and partnership is going to be a successful one. But given their current revenue flow and capital, they have an opportunity to be a market maker in terms of reaching out to so many types of partners and helping to develop markets and further entrench themselves as a part of that. So, they can help coordinate across different pieces of the ecosystem and start accelerating the development and movement in some of these areas. That being said, I think sometimes you hear a lot of talk about one topic and then the next GTC, maybe you never, you don’t hear a word about it. 

Patrick: Right.

Angela: But yeah, there’s definitely a give and take on success on those.

Ben: I would definitely agree with everything you said. NVIDIA being the market maker and really kind of making the ecosystems, as in kind of connecting partners that it has in different competencies, I see as a really core function of NVIDIA. I think even on their website, when I’m writing their quarterly report, I’m always kind of looking into developments with NVIDIA’s partner network. And there you can see how NVIDIA kind of segments its different partners into different groups. And I think in that sense, you get a real feel for how broad their partnerships are across different, kind of, industry verticals, if you will. So, for instance, I think one thing I see is NVIDIA partnering a lot, at least from Angela and I’s perspective, we follow infrastructure OEMs. So, we hear a lot of, kind of, partnership talk with the infrastructure OEMs, but then they’re bringing in, you know, GSIs from your lens. And I think together there’s room for a GSI to work in conjunction with NVIDIA and an infrastructure OEM in bringing a client a kind of comprehensive solution.

Patrick: Yeah, 100%. And I think that- I hope we’ll see that kind of go to market put together between, you know, sort of the three-way, four-way alliances. But one thing that sort of strikes me in what you’re saying is, and I guess I hadn’t thought about this until you brought it up, they do serve as an orchestrator within their ecosystem, which is different than most chip manufacturers, different than most manufacturers, period. We think of the consultancies and even the GSIs as providing that orchestration across the different technologies, bringing in the cloud and software and all that kind of stuff. Here NVIDIA is actually playing that role, which is kind of, it’s certainly unique and it certainly is in its own way kind of disruptive. But if they’re doing it as a way of, Angela, to your point, making the market, that is sort of a way of bringing everybody along. And if you have this sort of open system, developer-focused system in your DNA, then it’s easier to bring everybody along, right?

Angela: Absolutely. I think of them very much in this moment akin to what Microsoft has done in so many areas of the market where the technology is there, but they bring so many partners together. They help them innovate on developing features and capabilities that take advantage of the technology. So yeah, it’s really beneficial for all of the partners in the market in that sense.

Patrick: If NVIDIA becomes as ubiquitous as Microsoft, they will be a $15 trillion, $20 trillion company, right? Yeah, it is- it’s fascinating. And the automotive thing, I have to admit, it really caught my attention because it’s just such a, it’s so straightforward. Like, let’s just build the chassis that every robotaxi can be built on. It just makes complete sense. 

NVIDIA chip manufacturing in the U.S.

So, speaking of building things and manufacturing, Jensen said they’d be producing chips in the U.S. including the Blackwell within the next few months. And I say that because I can’t recall whether he said they’re going to start manufacturing now or start manufacturing in January or sometime later in the spring. But just to put it all in context, and we think back to where NVIDIA was just a couple of years ago when they were designing, but certainly not manufacturing. Is it a surprise that they’re going to manufacture themselves, that they’ve made that investment? And is it a surprise that they’re going to do it in the U.S.?

Ben: So personally, I don’t think it’s much of a surprise that they’re going to go to U.S. manufacturing. And I think it really revolves around what we’re seeing kind of through a political lens right now with the Trump administration. But even before that, with the groundwork laid with the CHIPS Act and the Biden administration, where it was really kind of a desire for the U.S. to reshore some of this manufacturing and control more of the chip supply chain. So, I know that TSMC has made really big investments being the most advanced chip manufacturer, expanding its fab capacity in Arizona. I want to say Samsung’s also increasing their fab capacities and with NVIDIA looking to continue manufacturing with these leading semiconductor fabs, it makes sense that they’re coming and kind of using that capacity that’s onshore in the United States. And I think without NVIDIA’s demand, and then also I think Apple’s a big player there, that it would have been hard to get TSMC and Samsung to make these investments and bring the manufacturing over to the United States.

Patrick: Right. And there was at one point, kind of, a very super geeky, in the details question about one little aspect of the manufacturing process and was NVIDIA going to do it or was TSMC going to do it? And there was this sort of back and forth of, you know, yes, TSMC is going to do it for a while, but we’re developing it. So, it sounds like it’s not just a one-off, let’s build a factory in Arizona. It’s going to be- it’s a big long investment by them and a strategy to actually build out that capacity, as you said, to sort of reshore that. So, Angela, any thoughts on that?

Angela: Well, as you describe that, it’s just, to me, I would call it, not to compare NVIDIA to all these different other giant companies, but to me, that’s very much the Apple play. Where, you know, in earlier iPhones, they relied on so many supply chain partners to build the devices, and over time, they have cut back more and more and, you know, figured out how to manufacture their own pieces there. So, I think strategically it helps you reduce costs likely. It helps you have more control over your supply chain, so eliminate risk. And if you’re in kind of that powerful position where you can develop those elements and make those investments, I think long-term, it could certainly be really beneficial.

Patrick: Yeah, and then by reducing your risks, certainly through supply chain, that frees up more of your capital to make those investments we were talking about before in your partners. So, yeah, it all makes sense. 

The competitive landscape for NVIDIA

So, then the next question, of course, is then when you look at the competition. Now, Ben, you sent me a note while I was down there because of somebody I was about to meet and you described how he was the guy who created the moat that NVIDIA enjoys being behind. I used that term with him, and he sort of bristled and said, it’s not a moat. Like, okay, I didn’t throw you under the bus and let him know that it was you that actually told me to say that. But in any case, they do have competitors. So, what do you think about the competitive landscape for NVIDIA right now? Are they in such a good place that they don’t need to concern themselves too much? Or are there threats to their arguably pretty dominant position in the market?

Ben: Sure. So, there’s a couple of things I want to unpack there. So, for those-

Patrick: First, a thank you for not throwing you under the bus. *laughs*

Ben: *laughs* Yes, first a thank you.

Patrick: Okay, there we go.

Ben: The man in question would otherwise be known as the father of CUDA and is really, I think, behind- and he won’t call it a moat. I think NVIDIA tries to toe a line between being seen as an open ecosystem player, but really being advantaged by some of its proprietary- the proprietary integration of how tightly its software and hardware work together.

Patrick: No doubt. Yup.

Ben: So, I think that’s given them kind of this crazy profitability that I always have to, I always have to check twice when I’m looking at a SEC filing. Those margins are wild. But I think where NVIDIA is advantaged and where they won’t really be impacted too much by competition in the near term is that they have a lot of these proprietary pieces in their integrated software and hardware stack. And I think we’ve seen NVIDIA with, for instance, the introduction of NVLink Fusion which allows for the networking between a third-party custom CPU, so something maybe like something developed by Amazon, Google, or Microsoft, to work with an NVIDIA accelerator or opposite for NVIDIA’s Grace CPU at the moment to work with the hyperscaler’s AI accelerators. I think by making that that announcement and bringing forward NVLink Fusion is kind of a way that NVIDIA’s strategically dismantling parts of its closed ecosystem moat in a way that probably sacrifices a little bit of lock-in and a little bit of profit margin in a way to maintain that dominant share. And I think they have so much margin to play with, and so many proprietary pieces in the stack that they’ll be able to slowly kind of dismantle those pieces and maintain market share. 

However, I will say when I think about who stands out as a challenger to NVIDIA, I see the biggest challenger today being AMD. While the hyperscalers do have their own custom AI accelerators, they’re not the same as an NVIDIA GPU or even what AMD’s offering with its line of Instinct AI accelerators in the sense that they’re not as programmable. So, they’re not as flexible to different workloads. And I think that’s where NVIDIA has been really advantaged is in the flexibility of their chips. Also, again, kind of going back to one of your questions earlier, you were talking about, you know, how NVIDIA is really the provider of all these developer tools. I think the developers are what drives NVIDIA’s advantage in the market. And the flexibility and tight integration of the software and the hardware is why NVIDIA will remain dominant, in my view, over some of these custom application-specific integrated circuits that are coming to market.

Patrick: Yeah, and they talked, Jensen in particular talked about the sort of virtuous cycle, and the developers were very much at the early part of that. And so just the virtuous AI cycle of generating the demand for their chips and then supplying the compute power and all that. So, and when you think about it in terms of AI is that there’s no one that stands out as like, this could be a- this could change things for NVIDIA if this particular company or solution or chip or whatever takes off specific to AI?

Ben: I see the biggest inhibitor for AMD is less related to the silicon because I think they’ve proved in some benchmarks that they’re roughly equivalent Instinct GPU to a Blackwell B200 could be more performant on certain workloads. The real differentiator is NVIDIA’s ecosystem of developers. And I think a way that that’s underscored is by kind of looking at what we see in enterprise AI. With NVIDIA AI Enterprise, their software stack for creating agentic AI solutions, the company charges I don’t know how many thousands of dollars per node on a subscription basis for access to that software. Where AMD is now trying to build a following of developers behind its ROCm platform, but AMD is giving away, AMD is taking the more open play, I guess you could say, but also kind of having to catch up and giving away their software for free.

Patrick: Right, which can work in the short term, but not in the long term, nothing can always be free. 

Ambition, risks and partnerships: Where will NVIDIA be a year from now?

So just to wrap up then, Angela, thoughts on where we think NVIDIA is going to be a year from now? Like what are some of the things you would anticipate coming from them, whether it’s new partnerships, possible acquisitions, changes in the way that they’re operating, pricing? What do you think is going to change in the next year with NVIDIA?

Angela: So, I think looking out over the next year, we’re going to see NVIDIA continue to be super ambitious. We talked about how beyond the data center, there’s expansion in so many edge, telecom, automotive, robotic areas. That I think again, we will be tracking how those particular areas evolve and to your point, whether or not NVIDIA is spreading themselves too thin, or if there’s areas that end up starting to take some precedence over others. Like telco, I think, is going to be a really interesting opportunity given just the challenges financially in that market for telecom providers. You know, can AI solve problems for them, or is it going to just mean more spending on infrastructure with capital they don’t have? I think that’ll be super interesting to watch unfold. And I think I’m interested to see, to some of Ben’s comments, how much we see the moat dismantle, the non-existent moat. And I think that I certainly expect more interoperability announcements and different elements of data center networking or the CPUs, GPUs, because there is a risk that customers will say, you know what, this is too much NVIDIA for me. And then that’s really the potential risk I see as far as the partnering and just general dominance.

Patrick: Yeah. Ben, how about yourself?

Ben: Going off that, I’d say the two biggest risks I see for NVIDIA, I think they’ll still continue those really large investments, which is supported by their growing partner network, but also just the incredible, kind of, top line growth that we continue to see with their chips. I think that- we’ll continue to see that just because of the rate at which the performance of each generation of GPU that NVIDIA is releasing delivers. But I think the two biggest threats to NVIDIA is really kind of looking at that coopetition lens where if you think about AWS, Google, and Microsoft as some of the company’s most prominent partners and biggest partners, they’re also ceding a lot of profit dollars to NVIDIA. So kind of building off that, the second threat that I really see kind of impacting NVIDIA and I think driving the dismantling of the moat is when one company has so much control of the stack, they’re making so much margin that I think that’s what really drives competition into the market. So, we’ll see more companies working with, you know, like the Broadcom’s of the world on XPU development.

Patrick: Right.

Ben: And I think that by taking those margins and for lack of a better word, being a little greedy there. I think that’s what’s going to drive investments from their competition. So, it will be interesting to see how NVIDIA toes the line between that kind of open and closed partner-led vs. not partner-led, how much profit they want to take. I think those are really the things that we’ll be watching in the next couple of years or even the next couple quarters.

Patrick: Yeah, it’s going to be fascinating. From my perspective, I’ll be watching very closely to see who coming out of that conference and going forward are their most active, vocal out there in the market, screaming about their partnership with NVIDIA, who among the companies that I cover in the services and the consulting space are increasing their efforts to piggyback on NVIDIA’s success. So, we’ll have a lot to cover. 

Final thoughts

Angela, Ben, thank you very much for coming in again. Appreciate this, and we will talk again soon. 

Angela: Thanks, Patrick.

Ben: Thanks.

Patrick: Tune in next week for another episode of TBR Talks. Don’t forget to send us your key intelligence questions on business strategy, ecosystems, and management consulting through the form in the show notes below. Visit tbri.com to learn how we help tech companies, large and small, answer these questions with the research, data, and analysis that my guests bring to this conversation every week. 

Once again, I’m your host, Patrick Heffernan, Principal Analyst at TBR. Thanks for joining us and see you next week.

TBR Talks: Decoding Strategies and Ecosystems of the Globe’s Top Tech Firms

Join TBR Principal Analyst Patrick Heffernan weekly for conversations on disruptions in the broader technology ecosystem and answers to key intelligence questions TBR analysts hear from executives and business unit leaders among top IT professional services firms, IT vendors, and telecom vendors and operators.

“TBR Talks” is available on all major podcast platforms. Subscribe today!