QCOM & AMZN Ink Multi-Generational AI Partnership Amid Rabid Compute Demand

SPEAKER_1 (0:00)

It's announced this morning for more details and what it means. Joining me right now, Daniel Newman, CEO of Futurum. I'm so glad you're with us because they're writing about it that it was the much needed good news for Qualcomm here. As we see here, Qualcomm is building these chips for Amazon. It's this partnership. Tell me about this.

Is this good news?

Daniel Newman (0:22)

Hey, Nicole, great to see you. Yes, I love days like this when, you know, we've spent a couple of years talking about Qualcomm's data center business, how this company can potentially become a force in this particular space. And obviously, as we've seen this market continue to grow in size, right? We went from hundreds of billions to trillions of dollars of TAM. We knew that Qualcomm needed a couple of anchor customers and needed a couple of these kinds of wins to legitimize itself. So being able to come out with a deal that has kind of a headline value of $60 billion of potential business. And of course, that's, I would say, not the ceiling if this is successful. But also has, it looks like about a $9 billion commitment based on the warrant structure of all of basic commitments that are already in hand. This is across their custom silicon, across networking. And Qualcomm and its ability to build low power certainly provides Amazon with a different toolset to go alongside its homegrown chips and of course, the merchant silicon that already uses. So, we see this is very optimistic.

It helps them hit that $15 billion target that they've been talking about for some time. And it certainly should get some of the skeptics that have doubted Qualcomm's ability to enter this space.

It should kind of at least quiet them for the time being.

SPEAKER_1 (1:42)

When we talk about AI ambitions and AI inference, because this is really part of the story, could you just explain that part of it more?

Daniel Newman (1:54)

Just to be sure, Qualcomm's AI ambitions?

I mean, this is a company that's entered the data center space.

It's on a few different occasions. But at this particular juncture, this was really a move for a company that had long been tightly coupled with handsets. We know that Apple had moved away from the business. That was something that some investors had a hard time understanding if Qualcomm was going to be able to fully recover. It did. But then Qualcomm moved into automotive, it moved into IoT. But Intel had some of this too, where it fell to the market that Qualcomm had not taken its expertise and its technology and fully maximized what's now a multi-trillion dollar TAM. We have the cumulative TAM for AI CapEx between now and 2030, up to $14 trillion, Nicole. They were in the space. And so they made some smart acquisitions. They've made acquisitions across software, across networking and across IP, to basically be able to build custom chips that can accelerate AI workloads, CPUs, kind of like Intel, AMD, ARM and others. They can make custom and off-the-shelf CPUs for AI. So they're building a diverse portfolio that companies like AWS, and here's the kicker, Nicole, it's not gonna be just Amazon. Amazon is the one you're hearing about right now, but this, I believe, is going to be the start of a number of these kinds of partnerships for Qualcomm, all of which powered by AI.

SPEAKER_1 (3:23)

Right. And they've made multiple partnerships, as you said, while they may have been relatively late to the AI story, but it does seem that Qualcomm has these AI ambitions because, as you said, they've been doing different partnerships, and I saw some with meta-platforms to use the Dragonfly C1000 processing units and Microsoft's Azure, for example.

So that's the big picture of where this is all headed, I guess, to a certain extent. What about Amazon Web Services? AWS is really a bright spot of Amazon, is it not?

Daniel Newman (3:56)

It is, of course. It's the largest hyperscale cloud by a margin. And this was another company that, in some ways, during the beginning of the AI boom, had to play a little bit of catch-up. Google and Microsoft had gotten out really fast.

But Amazon had been making multiple investments. It, of course, had the largest footprint of all the cloud providers. But it also has made investments. It built out its own CPU chips, its own custom AI chips. You'll hear about Graviton, and then you'll hear about Tranium.

And what you've seen over the last few quarters has been a real acceleration in their business.

They've seen their overall business catch up. The acceleration of the growth has been meaningful. And now they're looking for all the compute they can get. And that's been kind of the theme here, has been that nobody has enough compute. I had a number of inquiries today. You know, is this a sign of strength for Qualcomm? Is it a weakness for AWS? What's going on? And the thing is, it's strength across the board. It's that Amazon and AWS simply cannot build enough compute. It cannot buy enough NVIDIA, buy enough AMD and build enough of its own chips. And now they're looking for new, highly capable partners. And that's where Qualcomm showed up at the right time. So it's really a situation of and. In many cases, the market wants it to be or they want it to be, oh, this means less NVIDIA. Oh, this means less Graviton Tranium. We actually see it as all aggregate. The forecast continues to grow so rapidly for the amount of CapEx and the amount of AI compute required, that it's a bit of an all hands on deck. Qualcomm with its provenance and pedigree and low power, the acquisitions has been able to make is very complimentary to what Amazon is trying to do. And of course, Amazon is trying to ramp up compute quickly to meet its several hundred billion in backlog that sits in its cloud business today.