Remy Blaire: Taking a look at Oracle shares at this noon hour, they’re down a little over 5%, but are higher by over 4% this past week. Traders are expressing optimism about the company’s plans, with billionaire Larry Ellison’s company holding its investor day on Wednesday this week, where the cloud enterprise giant reaffirmed a target of $225 billion in total revenue by fiscal year 2030. The company also reconfirmed its fiscal year 2027 guidance of at least $90 billion in revenue, and in addressing concerns over escalating AI infrastructure spending, management laid out a clear timeline for capital intensity. CapEx is scheduled to peak in the next two cycles before stepping down in fiscal year 2029. And meanwhile, free cash flow is expected to turn positive in the second half of fiscal year 2028.
Well, joining me on this Friday to provide his outlook for Oracle following this week’s investor day is Ron Westfall, VP as well as practice leader for infrastructure and networking at HyperFRAME Research. Well, good afternoon, Ron. Thank you so much for joining us. It’s been quite a week here on Wall Street when it comes to tech, so I do want to zoom in on Oracle. Now, the company did reaffirm a massive acceleration toward $225 billion in revenue by fiscal year 2030. So given your analysis of enterprise infrastructure bottlenecks, do you believe supply chains can match this delivery curve before 2028?
Ron Westfall: I believe so. I think this underlines the fact that Oracle has firmly established itself as one of the top-tier hyperscalers in the market, and I think it’s demonstrating that they’re executing on their ability to meet the demand out there. In fact, it’s not just the demand, as we also saw that there is $664 billion in RPO that they have to pretty much convert into recognized revenue, and at least 80 billion that will be recognized in the upcoming revenue year.
And so I think that’s going to demonstrate that Oracle is on top of its game, but also that the concerns about an AI bubble are basically overstated. They’re overblown. And so what’s not to like about these data points? You know, having a fiscal year ‘30 projection of over a quarter-trillion dollars demonstrates that this demand is real, that the enterprises out there are already implementing AI across their organizations, both in terms of, say, conversational AI, but also agentic AI capabilities.
And this ongoing shift is tectonic. And this is something that’s going to be built into Oracle’s ability to not only meet this ongoing demand, but also demonstrate that this demand is quite simply tangible, and that it will continue to fuel the supply chain demand out there for data centers, as well as for Oracle capabilities that enable AI agent implementations across enterprises.
Remy Blaire: Yeah, and that’s an area that all of us are paying attention to. But building on what you just said, I do want to get your perspective on Oracle committing to capping incremental capital market financing at $20 billion. So how durable are these nontraditional funding structures?
Ron Westfall: Yeah, I think Oracle is playing it smart, in that they’re basically, you know, diversifying their capital raising and setting funding guardrails. So what that means is that they’re looking to raise up to $20 billion through fiscal year ‘30. So this is going to be distributed over the next few fiscal years, and this excludes refinancing of existing debt. That is something that is distinct and can take care of itself.
But I think what will be very interesting is that they’re splitting this capital structure across a $10 billion equity balance combined with $10 billion in gross debt. So really, I think the key takeaway here is that customer prepayments, as well as customers providing their own hardware, combined with vendor financing, will help soften these upfront cash outlays. That has been a concern for some of the market watchers out there. So this is, again, I think, demonstrating Oracle’s ability to adjust and adapt to, you know, the capital outlay concerns. And I think they have a smart plan in place, and they certainly have plenty of reasons to back up their ability to execute this. Just the RPOs alone will, I think, make a difference here.
Remy Blaire: And I do want to zoom in on the cloud. So how much of Oracle’s future growth will be autonomous cloud adoption, versus migration of legacy on-prem database workloads into partner clouds?
Ron Westfall: I think it’s going to be across the board. And yeah, that’s a very important question. Naturally, there will be demand for public cloud implementations for, you know, specific AI workloads. But I think what we’re going to see increasingly is the distribution of workloads becoming more implemented at the edge, requiring, I would say, not only smaller data centers, but also more edge support for AI inferencing.
And so what we’ve seen so far is a lot of focus on the AI training side of things, these massive GPU clusters that are being implemented across data centers nationwide and in other parts of the world, which are pretty much required in order to train the models that many of us use. It could be an OpenAI model. It can be a Copilot model. You get the whole idea.
But what is increasingly going to become very important is that now the models can adapt to real-time demands in terms of the inferencing. That is, you have a device, or, you know, a branch office, that is inputting a lot of new data, and now the models can respond to this dynamically in real time. And so this puts more emphasis on the hybrid approach toward cloud implementation. So you’re going to have both on-premises and public cloud implementations become increasingly combined. But there will be net new traffic more on the edge and less so in the purpose-built data centers out there.
So I think this is all going to play out to Oracle’s benefit and advantage, because they’re already a major player, as Oracle Cloud Infrastructure, on the hardware side. But they have all of these applications, for example their Fusion suite, that can also enable organizations to better adapt to and optimize AI inferencing at the edge. So this is, I think, what’s going to be unfolding over the next few years, and as a result, Oracle is strongly positioned to take advantage of this.
Remy Blaire: And Ron, while I have you here, I do want to get your take on what we’re seeing from the hyperscalers on this Friday. So we are wrapping up the month of October, and we are looking at a sell-off when it comes to the Nasdaq 100 and the Nasdaq Composite Index. And in particular, we’re looking at Alphabet lower and Intel lower following this week’s earnings. So what is your take on some of these names, and are you concerned right now?
Ron Westfall: That’s a very important question. I would say Intel is a very specific case, because they have a foundry business, so they have very unique requirements and also, I would say, a unique revenue situation. And so I’m not concerned about Intel so much in terms of the hyperscalers across the board. And I think, for example, with Alphabet, this is pretty much a hiccup. In other words, the hyperscalers have a very robust competitive situation ahead of them, even if there are, you know, days like this where there could be a combination of profit-taking and minor concerns related to other parts of the business that aren’t necessarily focused on the cloud side.
And I think this is reinforced by the fact that today, the capital expenditure across the hyperscalers is $800 billion. But what is underpinning that is the fact that the AI data center total addressable market is already $180 billion this year, and by 2030, that will be an $800 billion market. And that’s going to enable the hyperscalers to increase their capital expenditures even more, so that will be a $1 trillion business.
But it’s all reinforced by the fact that this demand is already in place, that they’re already looking at backlogs of over half a trillion dollars, $664 billion in Oracle’s specific case. But if you look at Microsoft, you look at Alphabet, and you look at AWS, it’s the same, I would say, fundamental principle: that this is simply a market that’s not drying up, that there is really no bubble on the horizon. And as a result, all the hyperscalers will continue to be strongly positioned to drive both the AI training and the AI inferencing requirements out there.
Remy Blaire: Well, Ron, we will have to leave it there for today, but thank you so much for weighing in. I appreciate all of your insights as well as your perspective.
Ron Westfall: Thank you.
