All right, good morning, everyone. Let's get on to the next presentation. We're very honored to have David Zinsner, the Chief Financial Officer of Intel, on stage. I'm going to read the safe harbor statement. It's second in its excitement to the risk factors in your 10Qs. That actually would be more exciting, believe it or not, a little inside joke. Before we begin, please note that today's discussions may contain forward-looking statements and are subject to various risks and uncertainties and may reference non-GAAP financial measures. Please refer to Intel's most recent earnings release and annual report on Form 10-K and other filings with the SEC for more information on the risk factors that could cause actual results to differ materially and additional information on our non-GAAP financial measures, including reconciliations where appropriate to the corresponding GAAP financial measures.
With that fun stuff out of the way, why does the government own 10% of you guys now?
Unless you've been living under a rock here, you probably saw the news. The government has put $8.9 billion into the company for about 433 million shares plus about $240 million in warrants. Look, you know we had about $5.7 billion of grants still outstanding. We also had received $2.2 billion. There were certain concerns around whether we would hit many of those milestones. They also have clawback rights on the $2.2 billion. Then there's the secure enclave money. There was a significant amount of uncertainty as to whether we'd receive that much or much of that cash. The government came forward, said, hey, rather than do this in the form of grants, what do you think about converting that to an equity holding? We saw a lot of advantages.
One, we were already thinking, hey, we likely will need a little bit more cash on the balance sheet, given that's one of Pat Gelsinger's core strategies is to solidify the balance sheet. This was a quick way to getting initially $5.7 billion in the door. By the way, we have received it. We got it last night. That's on the balance sheet. That was one thing. It eliminated the need to access capital markets in any other way in the near term. Also, like I said, given the uncertainty, this was effectively guaranteed that we'd get the cash. That was also pretty important to us. I think having the U.S. government invested in us and invested in our success is absolutely helpful. At the end of the day, we're going to have to bring out the processes and execute on the business.
To have their support and backing, I think, is helpful with customers. I think customers will acknowledge that takes us to a different level in terms of how they view us. We don't need to get, obviously, when you take the grant money and switch it over to equity, it does have a dilutive impact. We're getting a P&L benefit from the grants, and now we're issuing shares, which is also diluting us. When you run the math, it doesn't take a lot of volume of foundry wafers for us to make this accretive for us. We ran that math as we were doing the analysis and said, hey, I think based on this ownership, we will see that level of business going forward. Thus, ultimately, this should be pretty accretive to our existing shareholders.
I've heard various interpretations of the warrant side that you mentioned in the 51% threshold and all those sorts of things. Describe why that occurred, why that mechanism in the agreement.
I think initially, the government was thinking about this as some upside play for them. As we kind of worked the negotiations, we ultimately made the, it's a five-year warrant for what was roughly about 5% of the shares outstanding. We created a trigger that as long as we maintain a majority share of the foundry business, it would never trigger in that five-year period. Effectively, it reduced the cost of that warrant to something pretty nominal because we do have high confidence we're going to have this foundry business. We might, as we've said before, take outside investors into the foundry entity. We are structuring things in a way where we're separating ERP systems and creating a separate management and board structure over time for foundry. I don't think there's a high likelihood that we would take our stake below the 50%.
Ultimately, I would expect it to expire worthless. I think from the government's perspective, they were aligned with that. They didn't want to see us take the business and spin it off or sell it to somebody. In some ways, you could view this as a little bit of a friction to keep us from moving in a direction that I think ultimately the government would prefer we not move to.
Do you expect the U.S. government being involved changes the probabilities, all else being equal? We'll get into some of the other triggers that are necessary. All else being equal, improves the odds of your foundry business garnering external customers?
I think it's good endorsement. There's no question about that. Ultimately, most of what will make the foundry business successful is really in our hands. We've got to execute on the process. We've got to delight our customers. We've got to make sure we've got capacity there when they need it. Those things, I think, are going to be primarily the reasons why customers come to us. Endorsement helps on the margin, and I think ultimately will be helpful for us.
You also had a $2 billion investment from SoftBank. How did that come about? Was there any interaction or relationship between the U.S. government and SoftBank, those two moves that you made?
Yeah, it was coincidence that they fell all in the same week. We have been working for a number of years with SoftBank on ways to partner. We continue to have those conversations. They're very much invested in AI, as you know. We're looking for opportunities where we can intersect with that and provide solutions that either may be useful for them or may be in concert with them useful for others. The conversation evolved to, hey, we might be interested in making an equity stake. Ultimately, as we kind of work on the development of the relationship, SoftBank wanted to somewhat fast-track the investment. As I said, we were looking to put some cash on the balance sheet.
It was a good opportunity for us to do it quickly, get a good quantum of capital, a good slug of capital in one fell swoop with kind of a quality investor. I think we'll continue to have conversations with them about how we can partner and do things together that can be beneficial to both companies.
Is this likely to be the end of these sorts of capital raising efforts now that you've gotten whatever it is, $7 billion, $8 billion, I guess?
Yeah, I mean, this was actually a great quarter for us in terms of cash raise because we sold almost $1 billion worth of Mobileye. We're within a couple of weeks of closing Altera. That will generate $3.5 billion. We had the $2 billion from SoftBank and the $5.7 billion we just got from the U.S. government. We're in a good cash position, I would say. I feel pretty good about where the balance sheet sits. We're, as you know, in the process of deleveraging now. We had about $3.8 billion of bonds maturing this year. Some of it we kind of left the debt outstanding effectively by kind of converting it to commercial paper. Ultimately, by the end of the year, we'll be down $3.8 billion in debt. Our intention is to do the same with the debt maturing in 2026. I think we're in a pretty good place.
This is an expensive business, obviously. Depending on how successful the foundry is, there may be some capital needs as we kind of plan out our capacity build-out over time. I wouldn't think it happens until we have pretty significant demand on Intel 14A. As we talked about, there's likely going to be some opportunity for outside investors in foundry. That will probably be our second opportunity to raise cash to fund the growth on the foundry side.
Earlier this year, you and I talked about the fact that oftentimes you need a financial investor to set the value of a subsegment of your company before a strategic investor would come in. Has that already occurred with these two separate investments that we already discussed? I know it's in the company as a whole, not just the foundry side. Would we still have to have a kind of precursor be a financial before a strategic potential customer would put money into the foundry side?
This is in the foundry?
Yeah.
I don't know. I think we're open to both ways. It depends on the strategic and how comfortable they are at determining their own valuation for the business and whether that is relatively aligned with how we view the value of the business. Assuming that's the case, the likelihood is we would take a strategic investor without financial. We constantly talk to various firms out there about their level of interest on the foundry side, so I wouldn't dismiss the possibility that we might take investment from financial sponsors as well. I would just tell you we're probably a long way off from either of those things. We've got a lot of work to do to get the foundry business to a level where we feel confident it can operate somewhat quasi-autonomously. First, we've got to get through Intel 14A evaluation with customers.
We've got to ramp our Intel 18A successfully, which we're in the process of doing. Get through those things, and I think there may be an opportunity at some point to take some outside money into foundry. I think we're years away from that.
Let's pivot over to the technology side of things. We'll get to Intel 14A in a minute. Is success in ramping Intel 18A a prerequisite for Intel 14A working, or are they different enough that they're not completely codependent?
No, I mean, we're taking all the learnings of how, obviously, this was elongated in terms of our improvement on 18A. We would have liked to have gotten yield stabilized sooner. As we were adjusting performance, yield tends to be what gets impacted. We're in a really good place on the performance, and now we're making kind of steady incremental improvement on yields on 18A. We'll take those learnings to help us on 14A. There are differences. With 18A, it was a process that we intercepted early to make it a foundry node, whereas 14A, from the ground up, was always built to be a foundry node. The maturity of the PDKs is completely different in 14A. Just our level of rigor around the ecosystem is completely different versus 18A. It's already out of the gate looking better than where we were on 18A at the same time.
We'll port learnings. With every one of these processes, you learn a bunch of things in the previous processes that you take into the new one to help improve defect density and improve performance.
Is Panther Lake still on time?
Still on track. Yeah, things are looking good. Our first SKU will be out by the end of this year, and we'll have more SKUs in the first half of 2026. You'll really start to see the volume ramp as we kind of migrate through 2026.
On the 14A side, on last quarter's 10Q, you guys put in a risk statement, which we were joking last night nobody ever reads. This one actually was worth reading because you scared the pants off everyone. You basically said 14A, you need external customers, otherwise the return isn't there. Pat Gelsinger kind of summarized that even in the call. What led to that risk factor being for the first time inserted in your documents?
Yeah, I mean, look, the lawyers are always looking for areas where we should be elaborating in terms of our risks. The one thing about the SEC filings is you put all the risk potential corner case risk factors that could materialize. You don't put any of the mitigating factors. You don't put any of the things that might be upsides that might happen. I mean, that's just the nature of an SEC document. It's important not to read too much into a risk factor. We're constantly reviewing them and updating them. Pretty much every risk factor you read in there sounds horrible if it were ever to happen. It was interesting that this one got so much attention because a lot of them would be bad, obviously, if they materialize. I think more than anything, Pat wants to execute on 18A. He wants to execute on 14A.
He wants to maintain financial discipline while he's doing that. That's something, as he came in to be CEO, he really emphasized to the team. What we're trying to communicate is that the philosophy around why we're getting into the foundry business, you know, there's opportunity there. We think it's value creating for shareholders as well. The volumes required at these new nodes in terms of the spending level, it's hard to get an ROI unless you've got more volume than we have. We were able to do it with 18A. I think it'll turn out to be a good node for us from an ROI perspective without foundry customers. I do think that we have an opportunity in the second wave to get foundry customers into 18A. 14A is different. There's enough expense associated with 14A, investment associated with 14A.
The volumes for us, we are going to have to really emphasize 18A. We're going to be able to do that with the tile structure such that by the time we're ramping up volumes in 14A on our own, it's not going to be enough to drive an appropriate ROI for shareholders to make that investment. We do need customers outside of our own products business to drive volume through that business to make sure that we get the right ROI. We're just articulating that to the investors so they understand what we're trying to do. We're not going to put capital in place until we have firm commitments from customers that they're going to use 14A and that we will get the requisite volume that drives a good return. Now, Pat, I think if you ask him, every day he kicks the tires on 14A.
He becomes more and more confident around our ability to be successful there. The fact that Intel 18A, we already have made this point, but the fact that Intel 18A is now making steady improvement on the yields is also giving us a high degree of confidence that Intel 14A will be successful. It is not to say that we have any diminished confidence around Intel 14A. We are just going to maintain discipline around how we manage the build-out to be able to get the right return.
The fact that Pat Gelsinger talked about it so proactively on the call, the risk factors and everything, and you put it in the 10Q, and I know all the caveats that you just said that you don't get to say the mitigating side of the equation. I've heard people think that or kind of posit the idea that you could scare potential customers away, or you could kind of get them off the sidelines that they have to step up now if they ever want you. Otherwise, they're going to deal with a monopoly. Which one ended up happening?
I’d say it this way. I think customers, for a couple of reasons, will get real value out of us as a foundry. Number one, it helps diversify their supply chain, which is obviously helpful. More importantly, it's always good to have multiple suppliers to help on the pricing side. Every customer or every supplier brings something unique, obviously, to the table. I think that will be the case for us with Intel 14A. There will be opportunities where that is the differentiated technology that certain customers will require. That’s how they'll make the evaluation, and that, to me, is how we'll win the business. On top of that, they obviously have to get comfortable that our process is a good process. We're in the early stages with customers, working with them on the data.
We didn't announce when we expect a 0.5 PDK, but it's coming in the not-too-distant future. That will be a key milestone for customers to evaluate our technology. To me, I don't think anything in our risk factor is really driving customers one way or the other. It's what we're doing on the field in terms of supplying them the technology that they need at a price that works for them that will ultimately determine whether we're successful there with customers.
On the 18A side of things, the way you guys structured the technology was very kind of high-performance compute-friendly and a little less kind of mobile-friendly. Does the same thing apply to 14A? I know there's different flavors of 18A, but just kind of at the highest of levels.
Yeah, look, I think with backside power in our solution, it definitely gives certain customers an advantage. That generally is in the high-performance compute. That is the area that we're generally targeting. That said, I think the mobile or handset area is an opportunity for us. We're not going to win every customer, obviously. The incumbent is a machine in terms of their ability to execute. What we're looking for is just we don't need a ton of volume from customers. Most of them have some form of high-performance compute in their portfolio that could be a pretty interesting opportunity for us to win. I wouldn't dismiss handset. There could very well be opportunities in the handset space to win as well.
Fourteen A customers are going to happen. They'll probably not be named by you. If you were to win a foundry customer for Intel 14A, what is the timeline necessary as you try to balance the ROI and the investments and all those sorts of things?
Yeah, I think next year is going to be an interesting year for us. We'll see whether we get to a point where we have the ability to announce a win. Lip-Bu has been also, when he came in, he emphasized this. He's not declaring a customer win until it's a customer win with a real commitment and signature behind it. All the pipeline stuff, you know, obviously, we track internally. We're not declaring anything until we've got somebody signed on the dotted line. I expect next year will be a good year to evaluate us. If we don't win a big customer next year, I don't think that takes us out of the window of opportunities for Intel 14A. I think 2027, we could win customers as well. That still would get us a good business on Intel 14A.
Whether it's an external customer or an internal Intel ramp on Intel 14A, is the time frame kind of 2028, 2029, 2030?
I think most, those in the foundry space competing at that node are generally talking about this as a 2028, 2029 type introduction. Obviously, the expectation would be it would last a fairly long time after that. That kind of very late in the decade is probably when you'll start to see volume in the leading edge customers.
Gotcha. Why don't we talk a little bit about the product roadmap side of things? So much of what we've discussed so far has been on the Intel foundry side. If we pivot over to the product side of things, talk about between notebook, desktop, and server, where you think the product line is strong or needs some work. What are you guys doing to address it?
Yeah, I think on notebook, we're in pretty good shape. We're now in the process. Meteor Lake was our product kind of last year. We introduced Lunar Lake this year. It's in the process of gaining adoption. We're expecting this quarter to be a pretty good quarter for Lunar Lake. Notebook, I think, is good, solid. Share is solid. As you know, we kind of fumbled the football on the desktop side, particularly the high-performance desktop side. As you kind of look at share on a dollar basis versus a unit basis, we don't perform as well. It's mostly because of this high-end desktop business that we didn't have a good offering this year. Nova Lake, which is the next product, is a more complete set of SKUs. It does address the high-end desktop market. We would expect that we will improve our position next year.
All in all, I actually feel pretty good about the client. It's not executing flawlessly, but it's executing pretty well. That's our business that generates good operating margin, drives good cash flow, and funds the rest of the business. On the data center side, it's been pretty mixed. We are continuing to bring out products that do perform better than the prior product. We're still not there relative to the competition. We're really performing well in areas where there's a single-threaded solution. We're performing well in what they call the head node, which is a CPU that runs alongside the GPU. Those areas are bright spots. Collectively, across the entire data center, we still have more work to do to get products that are really performant across the spectrum and are meeting our customers' expectations. We were already on a path towards doing that.
I would say, as Pat Gelsinger has come in, he has really rolled up his sleeves and torn apart that strategy and looked at the roadmap. There were gaps, quite honestly, in that roadmap that we were allowing that Pat Gelsinger is not going to allow, particularly around multi-threading. We will be adjusting the roadmap to make sure that we are listening to customers and delivering products that customers want and need. That takes a bit of time. I think we'll make some incremental improvement over the course of the next couple of years. It's going to be a multi-year process to get the portfolio to be really where we want it to be.
Is Diamond Rapids something that you hope closes the gap? It's better than Granite, but it doesn't get you to.
It doesn't get us quite there. Yeah, I mean, it does in certain cases. The performance is actually better, but in other cases, it's not. We've got more work to do to finally get to a place. It's really not. I think Lip-Bu actually named the product in some forum, but Coral Rapids is the next product. That's our real opportunity, I think, to begin to take a really good step forward.
If we think across those products just to kind of blend the product and the foundry side, talk a little bit about the external to internalization of where those are produced. You know, you talked about Panther versus Lunar and then Nova and then the same thing on Diamond versus Granite, et cetera.
Yeah, so obviously Panther Lake, that's an 18A process. A lot of wafers coming back for that. Generally speaking, our data center products are done internally to help support foundry. The great thing about just in general where the architecture is going is it's moving more and more to these tile architectures. That allows us to kind of pick and choose. In a lot of cases, our Intel Foundry Services business will be the best combination of price and performance. In other cases, external foundry will be the right balance. I think the foundry business has a great opportunity to continue to grow the wafers as they move to 18A and 14A. Prices get better. Margins get better. They're really in a good place. We do want products to have the flexibility to pick the right silicon for what architecture they need.
Likely, for sure, the majority of our wafers will come from Intel Foundry Services on the product side. They will continue to be a significant purchaser of external wafers for solutions that require it.
Let's talk a little bit about the CapEx side of things. You guys, I think, have alluded to next year will come down year over year. I know there's a gross versus a net dynamic to think about. Now we have change in what were grants from the government and now equity investments that you won't get next year. Just talk a little bit about how you're thinking of CapEx and kind of where is the normalized level Intel can operate on whichever one you want, gross or net.
Yeah, mostly looking at it on a gross basis. I think where we are here in this roughly kind of $18 billion level is, I think, a pretty good level for the next few years. We have the advantage at this point. We spent a lot in prior years, and we built up what we call our assets under construction, work in progress, CapEx that we're digesting now. That is enabling us to deploy more capital than the $18 billion on an annual basis, but not spend more than $18 billion because we can leverage what we've already purchased. Naga and Lip-Bu have been pushing on reuse too. I don't think we did a particularly good job on reuse. We can do better.
They are really forcing the organization to not dispose of equipment and sell it at pennies on the dollar, but reuse it in the process rather than going to something brand spanking new on the equipment side. That is also helping us keep the $18 billion check in check. Lastly, just in general, we were kind of out of benchmark with when we bought equipment and brought it in and worked our way to getting it in service. That time span versus what our peers do was way more elongated. We have tightened that up to be more efficient in terms of when capital comes in and the time frame from that point to the time it goes into production. Squeezing that time period has been a big measure for us more recently. We still think, though, that next year we can bring the CapEx in a bit.
The question that got asked in the call was around maintenance capital, and that question kind of came in and somewhat, I think, confused the conversation a bit. I think we'll be down in CapEx, but I don't think it's going to go to $9 billion or anything close to $9 billion. It's going to be somewhere in the teens, a couple of billion dollars below. $18 billion is probably the best we could probably pull off. After that, I think it's normalized to this kind of roughly in this high teens level for a couple of years. We see what we need in terms of equipping out Intel 14A. We will have a high-class problem if we do win a significant slug of the foundry wafers on Intel 14A because we will have to invest for that. That is likely to push the CapEx back up.
I think hopefully we are making a good amount of progress on the product side in terms of improving revenue and margins there, and that should generate cash flow, which should help support those CapEx investments.
Let's talk about gross margin a little bit. You guys have kind of been around the 40% mark, plus or minus. You have a target that's significantly above that. We can talk about how you get there in a bit. In the nearer term, say the next year or two, what are the pluses and minuses to getting above the 40%? Are you guys trying to fix so many things at once that operating at this level is not your goal, but you're kind of fine with it because there's bigger things to address?
Yeah, no, we definitely want to get back to a four handle. Obviously, we've quoted much higher numbers per our model. By the way, a good running business should be more in those model numbers. We've got a lot of work to do in the near term just to get the business up above 40%. I think on the foundry side, we're in pretty good shape. Like I said, Intel 18A and Intel 14A have very good cost structures relative to older nodes. The pricing is better. They enjoy that pricing from the products business. Also, as we win foundry customers, we'll get that price as well. They should be on a steady path to improving both gross and operating margins as we go into 2026 and beyond. I see that as a pretty good tailwind.
I think on the products business, we've got to get our portfolio to a place where we're getting the right ASP, really. That will be the single biggest, I think, driver of gross margins on the products business is to bring out. Nova Lake in the client side, great opportunity for us. As we progress from Diamond Rapids to Coral Rapids and so forth, great opportunity for us to command better pricing as we deliver more per lot to our customers. We still have a lot of work to do on the cost side, too, quite honestly. That's a lot more in our control in the near term is to kind of chip away at our costs. Some of it takes longer, like, for example, how much silicon is being used in the design. I don't think we were very efficient in terms of how we used silicon.
Packaging, we've tended not to be as focused as we should be in terms of our use of advanced packaging. We rarely focused on test times in the past. That's become a real focus of the products business. They spent a lot in terms of samples. Of course, you want to give samples. I don't think they were really paying attention to how much they were giving away in terms of sample activity. They've kind of looked at that and tried to be more efficient around that. Lastly, just the throwaway cost of inventory reserves. We've had a lot of inventory reserves over the last couple of years. We can do a lot better. I think there's a lot of opportunity over the next 12 - 18 months to improve gross margins on the product side. It might not come in the first few quarters.
I do expect that to be a tailwind. Obviously, as I said, foundry should start to see margins improve next year, which will help as well.
Ultimately, Equal should be knocking wood past the trough in gross margin?
I would think so, yeah.
In the last minute or so we have left, I just want to wrap things up with Pat coming back. He was on the board for a period, left, and now came back. You saying it in Intel, both of you guys are industry veterans, highly respected. You're not taking on an easy job turning around the company. Clearly, you see something that keeps you excited about it. Why don't you just summarize in a minute, literally, that we have left, what's the vision you guys have?
Don't answer anything in a minute, Ron.
Yeah, where Intel could be.
Yeah, look, I think there's a vision out there. We get this foundry business to be successful. A successful foundry business can trade at multiples of net book value, one measure of it. We have a lot of net book value in our foundry business, so there's a real opportunity. I think we're getting virtually nothing for it at this point in terms of our valuation. There's a great opportunity to create shareholder value through making this successful. It's going to take some time, as I talked about. I see the opportunity. With Pat coming in and the talent he's recruiting, I feel really good about the opportunities there. On the products business, this is just blocking and tackling. We've just got to do better in terms of executing on the product portfolio. We've got to listen to customers more, and we've got to be disciplined around our cost structure.
We do those things, and the valuation there should be significantly above where we are. That's the thesis. I'm excited for Pat joining, and I've already seen some real changes at the company that lead me to conclude that we have a real opportunity to drive some significant shareholder value over the course of the next four or five years.
Perfect. Dave, we truly appreciate you coming down to the Deutsche Bank Tech Conference and spending some time with us. Thank you very much.
Thanks for having me up there. Thanks.