Intel Reports Second-Quarter 2025 Financial Results

[Y.H]

If 14A doesnt get committed volume in the next 1 year. 18A is the last node. We have seen this story before and we know how it ends.

Sorry but how would that ends ?

 

[benb]

It's a very easy path for TSMC to make Intel an offer they can't refuse on 14A, converting a large customer to the vast, unbeatable economy of TSMC scale. In contrast to the contracting, declining economy of Intel scale. And do the same with any wavering large customers who might defect to Intel or Samsung.

The AI winner has already been announced, and it is not Intel or Samsung. The economics of scale just get worse for Intel (and Samsung) from here. TSMC wins it all. So it goes.

 

[hist78]

https://www.intc.com/filings-reports/all-sec-filings/content/0000050863-25-000109/0000050863-25-000109.pdf

"As part of the transformation of the company, we have begun implementing a more disciplined approach to the deployment of capital. The design, development, and manufacturing of leading-edge semiconductor manufacturing process technologies, or nodes, is risky and capital-intensive, and it takes years for capital investments to yield a return. Under our more disciplined approach, we intend to invest capital in future node development and additional or upgraded manufacturing facilities only where we have a clear line of sight to an acceptable return on that capital. We expect to release the first SKU of our first products manufactured on our new leading-edge node, Intel 18A, by the end of 2025, and continue to develop its derivative node, Intel 18A-P, designed for future Intel products and external customers.

We are focused on the continued development of Intel 14A, the next generation node beyond Intel 18A and Intel 18A-P, and on securing a significant external customer for such node. However, if we are unable to secure a significant external customer and meet important customer milestones for Intel 14A, we face the prospect that it will not be economical to develop and manufacture Intel 14A and successor leading-edge nodes on a go-forward basis. In such event, we may pause or discontinue our pursuit of Intel 14A and successor nodes and various of our manufacturing expansion projects.

While we continue to evaluate Intel 14A for use in future Intel products and our plan includes an initial product designed to utilize Intel 14A, at present we are maintaining the option to design future Intel products requiring nodes with performance beyond Intel 18A and Intel 18A-P to be produced internally or by an external foundry. If we were to discontinue development of Intel 14A and successor nodes, we expect that a majority of our products would continue to be manufactured in our own facilities utilizing our nodes up to Intel 18A-P through at least 2030. By focusing on our customers and delivering the best semiconductor products to the market, manufactured on the most appropriate internal or external node from a performance and cost perspective, and only deploying capital on new nodes and manufacturing facilities where we believe they will yield an attractive return, we believe we can improve the competitiveness of our products business, and the overall financial results for the company."

Intel CEO Lip-Bu Tan responded to a UBS analyst's question about the possibility of 14A being canceled. He stated that unless he sees sufficient orders from both internal Intel teams and external customers, he will not approve Capex allocation for 14A.

This suggests that 14A has only one to two years to prove its relevance due to the long lead time required to build a new fab. Major fabless clients and even startups are unlikely to risk their money or place their dreams on Intel Foundry unless Intel can demonstrate firm commitment.

Did Mr. Lip-Bu Tan just reveal the unthinkable that Intel might halt development of 14A?

 

[MKWVentures]

Sorry but how would that ends ?

You say you have a great next node. No one wants it. you cannot fund it so you stop work (you buy chips from a different foundry). You write papers how it is a awesome node. You try to license it to someone else.... they dont want it even if it is free. IBM2.0. The difference? IBM actually had a reasonable large foundry business before giving it away. Intel hired the people that saw the end of IBM foundry and then GF deciding to stop work on advanced nodes. They can help.

 

[Y.H]

You say you have a great next node. No one wants it. you cannot fund it so you stop work (you buy chips from a different foundry). You write papers how it is a awesome node. You try to license it to someone else.... they dont want it even if it is free. IBM2.0. The difference? IBM actually had a reasonable large foundry business before giving it away. Intel hired the people that saw the end of IBM foundry and then GF deciding to stop work on advanced nodes. They can help.

So if 18A is the last node, will a break up still happens?

 

[MKWVentures]

So if 18A is the last node, will a break up still happens?

I am sure they are looking at all options. We will publish some likely scenarios on our website

 

[hist78]

You say you have a great next node. No one wants it. you cannot fund it so you stop work (you buy chips from a different foundry). You write papers how it is a awesome node. You try to license it to someone else.... they dont want it even if it is free. IBM2.0. The difference? IBM actually had a reasonable large foundry business before giving it away. Intel hired the people that saw the end of IBM foundry and then GF deciding to stop work on advanced nodes. They can help.

IBM’s semiconductor story is far from over. After a legal battle in which IBM and GlobalFoundries sued each other for breach of contract, the two companies reached a settlement in January 2025. Meanwhile, IBM continues to roll out its own server CPUs, designed inhouse and manufactured by Samsung.

IBM also licensed its 2nm technology to Japan’s Rapidus, aiming for mass production by 2027. In addition, IBM remains active in semiconductor research at its Albany, New York research lab, backed by funding from the CHIPS Act and the State of New York.

The IBM semiconductor saga continues...

 

[Y.H]

I am sure they are looking at all options. We will publish some likely scenarios on our website

I'd always had this feeling that except for maybe hyoerscalers and Broadcom, there wouldn't be much business from everyone else. Certainly not AMD or Nvidia ( despite what they said previously). This is because Intel is still competing with them. Why give them business to complete with you in other areas. It's always better to starve your competitor than you feed them. If second source is important,let some one else spend the money. I see this kind of thinking frequently in industries I'm familiar with all the time. I don't work in semiconductor but I am guessing similar thinking should happen. The desire to starve your competitor or more likely the fear that you are helping your competitor to compete with you often triumph over all reasoning.
It is in this light that I think maybe a break up would work and maybe business may pick up significantly after that.

 

[blueone]

Simple question... Right now, who would you trust to deliver foundry wafers in the next 4 years. TSMC or Intel?

TSMC, of course, but not if Taiwan is invaded. I trust China to execute on a threat more than I do the US (like about Greenland). It just seems like the world really needs a back-up plan. I can't escape the feeling that it's a smart thing to do, and not just for the US.

 

[Daniel Nenni]

Did Mr. Lip-Bu Tan just reveal the unthinkable that Intel might halt development of 14A?

I believe he did, which is a very honest answer. He clearly said it was not a Field of Dreams situation where you build capacity and customers come. TSMC does not do that. TSMC get's pre pays and wafer agreements then builds capacity which is what LBT will do. CC Wei always says capacity is tight because TSMC builds on customer orders not pipe dreams of being the best so customers will come crawling.

Whether Intel continues in the foundry business depends on customers wanting an alternative to TSMC. I'm sure Lip-Bu is hard selling companies like Apple and Qualcomm to be manufacturing partners like Apple and TSMC did at 20nm. If he is not successful Intel cannot afford to continue to be a foundry, they will go back to IDM 1.0 of fab-light.

Please remember, if not for Intel we may not have HKMG, FinFETs, BSPD, and many other semiconductor manufacturing innovations. TSMC would also not be in the place they are today. How fast will TSMC innovate without a leading edge competitor? This is a much more serious situation than most people realize. Lip-Bu Tan is opposite of Pat Gelsinger. Lip-Bu laid his cards on the table and now we will see who recognizes how important Intel is to the United States and the semiconductor industry.

 

[siliconbruh999]

I am very happy with cancellation of German Expansion should never have been started

 

[Y.H]

I believe he did, which is a very honest answer. He clearly said it was not a Field of Dreams situation where you build capacity and customers come. TSMC does not do that. TSMC get's pre pays and wafer agreements then builds capacity which is what LBT will do. CC Wei always says capacity is tight because TSMC builds on customer orders not pipe dreams of being the best so customers will come crawling.

Whether Intel continues in the foundry business depends on customers wanting an alternative to TSMC. I'm sure Lip-Bu is hard selling companies like Apple and Qualcomm to be manufacturing partners like Apple and TSMC did at 20nm. If he is not successful Intel cannot afford to continue to be a foundry, they will go back to IDM 1.0 of fab-light.

Please remember, if not for Intel we may not have HKMG, FinFETs, BSPD, and many other semiconductor manufacturing innovations. TSMC would also not be in the place they are today. How fast will TSMC innovate without a leading edge competitor? This is a much more serious situation than most people realize. Lip-Bu Tan is opposite of Pat Gelsinger. Lip-Bu laid his cards on the table and now we will see who recognizes how important Intel is to the United States and the semiconductor industry.

I suspect Broadcom might be a good potential customer. I don't think AMD and Nvidia would be interested.

 

[siliconbruh999]

My top Picks are Mediatek two of the big cloud vendors Qcom.
If they get any part of apple business that would be good enough but it's a pipe dream

 

[Brady]

Please remember, if not for Intel we may not have HKMG, FinFETs, BSPD, and many other semiconductor manufacturing innovations. TSMC would also not be in the place they are today. How fast will TSMC innovate without a leading edge competitor? This is a much more serious situation than most people realize. Lip-Bu Tan is opposite of Pat Gelsinger. Lip-Bu laid his cards on the table and now we will see who recognizes how important Intel is to the United States and the semiconductor industry.

Customers really do need to step up, and LBT said it straight up: join us in designing and making 14A a successful node -- like you already do with TSMC -- or all you'll have is a monopoly to work with going forward (Samsung notwithstanding...).

I do love the transparency. Gelsinger gave Foundry the money to step up on and become viable on the leading edge, but was a lot of "rah rah" and bluster. LBT is bringing the ruthless business acumen.

We'll see what leading edge customers choose... a monopoly or a competitive market.

 

[Y.H]

Customers really do need to step up, and LBT said it straight up: join us in designing and making 14A a successful node -- like you already do with TSMC -- or all you'll have is a monopoly to work with going forward (Samsung notwithstanding...).

I do love the transparency. Gelsinger gave Foundry the money to step up on and become viable on the leading edge, but was a lot of "rah rah" and bluster. LBT is bringing the ruthless business acumen.

We'll see what leading edge customers choose... a monopoly or a competitive market.

Honestly, do customers truly care about having a monopoly? It sure seems like it's a good statement to put out but it don't actually translate to any action

 

[Daniel Nenni]

I suspect Broadcom might be a good potential customer. I don't think AMD and Nvidia would be interested.

Could be, Lip-Bu and Hock Tan run in the same circles. MediaTek would be interesting. The CEO is Rick Tsai who was tossed from TSMC by Morris Chang. I was actually shocked when MediaTek partnered with Intel Foundry due to the Taiwan connection but if Qualcomm is getting preference at TSMC MediaTek could certainly do a partnership deal. MediaTek's market CAP is under $80M though so they are not a whale really. Broadcom's market cap is $1T, Qualcomm is $175M, Marvell is $60B, another possibility. Apple would certainly be the best company to fund Intel and teach it how to be a leading edge foundry. Intel could also have some new technology that Apple could leverage, anything is possible here.

AMD or Nvidia is a longshot. The Taiwan connection is really strong and Intel competes with them. Exciting times in the semiconductor industry!

 

[Y.H]

Could be, Lip-Bu and Hock Tan run in the same circles. MediaTek would be interesting. The CEO is Rick Tsai who was tossed from TSMC by Morris Chang. I was actually shocked when MediaTek partnered with Intel Foundry due to the Taiwan connection but if Qualcomm is getting preference at TSMC MediaTek could certainly do a partnership deal. MediaTek's market CAP is under $80M though so they are not a whale really. Broadcom's market cap is $1T, Qualcomm is $175M, Marvell is $60B, another possibility. Apple would certainly be the best company to fund Intel and teach it how to be a leading edge foundry. Intel could also have some new technology that Apple could leverage, anything is possible here.

AMD or Nvidia is a longshot. The Taiwan connection is really strong and Intel competes with them. Exciting times in the semiconductor industry!

Dan is one of the few optimists thus far. So far, from most comment I read, everyone seems somewhat downcast with the potential end of leading edge manufacturing in US.

 

[Daniel Nenni]

Honestly, do customers truly care about having a monopoly? It sure seems like it's a good statement to put out but it don't actually translate to any action

If a customer thought they could have a competitive edge by working with Intel Foundry they would. But you are right, altruistic they are not. My optimism is based on Lip-Bu. I have complete confidence in him.

 

[siliconbruh999]

I am preparing for 18nm+++++ btw since he said 18A will be a looooong node

 

[hist78]

Dan is one of the few optimists thus far. So far, from most comment I read, everyone seems somewhat downcast with the potential end of leading edge manufacturing in US.

TSMC’s Arizona fab is producing “4nm” semiconductors, part of the N5/N4 family. By most standards, this qualifies as a leading edge node. Unless we define “leading edge” strictly as the single most advanced node in the world at any given time. But that definition constantly shifts due to rapid technological progress, making such a pursuit less meaningful and harder to sustain.