welcome to Chip War the first jointly produced webinar by Intel Alumni network and Intel Corporation I'm happy to welcome CEO Pat gelsinger to kick off today's event Pat you have the floor hey thank you thank you Carl and great to see you great to have those in the Alumni network joining us and a super interesting topic and about uh you know uh 38 years ago I think it was uh when we were just starting up the 386 project Carl's old enough to remember that but I remember at the time that uh you know Jim slager who some of you might remember you know comes walking into my office and hands me this Pulitzer prize-winning book uh at the time about the design battle a data General that was going on and he hands me the book and he says read it you're living it right and I remember at the time it's from like wow I didn't realize it and you know all the things that were going on in 386 the 432 what became the 960 project you know but a book that was chronologizing the environment that we found ourselves in and the Book of course Tracy kidders a great book at the time and one that I quite enjoyed uh you know for many years thereafter when I first picked up the book chipwar by Chris Miller that exact thought crossed my mind read it to everybody at Intel because we are living it you know we're in the middle of this seminal period of History where where we're rebuilding the technology Supply chains of the world and as I've said where the oil reserves are has defined geopolitics for the last five decades where the technology Supply chains are is more important for the next five decades let's build them where we want them and as someone in that context that I think Chris is and Hill you know certainly give you in his own words you know sort of he didn't start out writing ship war right but the understanding of how Chip War you know came about and how it chronologizes the world that we at Intel as we're rebuilding not just the company rebuilding is position in the technology industry but also rebuilding the technology Supply chains of the world and I think you're going to Super enjoy the conversation today Chris and I have done a couple of in interviews together over time and uh you know have really benefited from you know his research and work and I'm looking forward to Chip War part two and one that we're looking forward to him writing in a few years so back to you Carl and thank you for letting me join you all today thanks Pat I really appreciate your uh comments and your uh taking time to start the event off today I know it means everything to all the Intel employees as well as the Intel alumni so um thank you much and we'll see you again soon uh please note everyone that we are now live worldwide with participants from Intel Alumni network and Intel employees before we get started let me cover some housekeeping items first on questions and answers Intel employees using your smartphone please scan the QR code in the lower right hand corner of the screen to submit your questions Intel alumni please use the same process we will go to audience questions immediately after my kickoff round of questions for each of our panel members concludes if you are unable to attend or want to review the proceedings Intel will post an available link later this week on the internal Intel website or Intel alums it will be available at www.intelalumni.org it is certainly my pleasure to host the event today and I would especially like to thank Intel and Intel Alumni network people who have worked closely together to produce this first of a Kind joint online event that's truly great example of risk taking it is my pleasure to host the event today and I would especially like to thank Intel and the Intel Alumni network people who have worked closely together produce this online event and from Intel that would be Max tories Kimberly samarjia sumitra sukumaran Danielle Goodwin Penny Bevins and from Intel Alumni network many thanks to James cape and the ien board of directors for continuing their efforts to produce engaging in thoughtful online events I really appreciate the time and effort our panelists put in in preparing for the event it is certainly a topic worthy of discussion please welcome Jim Evers vice president Ohio site manufacturing and operations manager stupan senior vice president and general manager of Intel Foundry services and Al Thompson vice president of Intel government affairs our special guest today is Chris Miller author of Chip War Chris thank you for your time today you have written an outstanding book on the history and the future of the semiconductor industry I enjoyed reading it and recommend everybody buy a copy and give it a read only once in a long while does such a book come along that is so timely and has so many important strategic implications thank you for highlighting and educating us all on the strategic competition we are facing today not only is the mantle of Industry leadership at stake going forward the very security of Nations is in the balance it reminds me of two books and their authors that had great impact on Intel strategy during my time at the company Michael Porter's competitive strategy and Clayton christensen's the innovators dilemma Andy Grove has these two books around to executive staff members with the request that we think about our business responsibilities within the models presented by these two authors Chip War represents a similar opportunity for reflection on Intel strategy the next era of competition the semiconductor industry will bring an old procurement Ally Into The Fray governments around the world will insert themselves as never before they have always been involved but now there is consensus on the importance of semiconductors economically technologically and for National Security I think if I think about the model implied by Chris's work governments will have to step up their involvement to a new level of partnership with Intel at the same time the U.S government will need to engage in reforms that focus on technical innovation while reforming the federal procurement process I am pleased Now to turn over the floor to Chris Miller for his opening remarks to start us off today Chris you have the floor well thank you very much Carl for that very generous introduction thank you Pat for kicking off this conversation and thanks to the entire Intel and Intel alumni a team for the invitation to join you all today when I was researching my book I had the great honor of speaking with many internal alumni who are scattered all over the semiconductor industry who actually defined in many ways the history of the semiconductor industry and so I've learned much from speaking with many of your colleagues over the past several years as I researched and wrote chip or what I'd like to suggest today first off looking historically is that Intel is the most important company of the last 50 years and I say that not only because I know there are many current and former employees of intel on this call but because I think it's historically true and second what I'd like to suggest to you is that the future of the capabilities that Intel has played a central role developing of computing power that will be critical to the balance of economic technological and also security over the coming half century now it might seem like a somewhat grandiose claim uh to argue that Intel is the most important company of the last 50 years but I think I can fairly straightforwardly convince you that it's true because the most transformative Force across the entire world over the past has since half century has undeniably been the development of Moore's Law which of course as all of you know is not actually a law of nature but rather a prediction made by one of the co-founders of Intel Gordon Moore now over half a century ago almost six years ago and Moore's law has persisted uh in no small or thing to Intel which has been one of the companies that has invested uh very heavily over the past half century in driving Moore's Law forward and within the semiconductor industry Moore's Law is often taken for granted because people like you know that it's been accomplished and know what it's taken to push it forward but in the context of the rest of the economy and indeed the rest of all of human life Moore's Law is simply unprecedented there's nothing else in the history of Economics that has managed to double every two years for such a long duration of time nothing else comes remotely close in most businesses if you grow your productivity by a couple percentage points a year you've accomplished success economies in aggregate grow by a couple percentage points a year total Factor productivity across the U.S economy Grows by one two or three percentage points a year depending on the year and yet the chip industry has managed to double the number of transistors per chip and thus the Computing capabilities per chip almost every other year now for over half a century and if you think of the painfully slow rates of improvement of all other types of Technology the extraordinary things that the chip industry has accomplished begin to become clearer and to put this into context I like to think about the aviation industry which on the one hand is an example of pretty cool high-tech stuff planes can fly that manages to Astound my uh three-year-old son on a regular basis and indeed when we're sitting on airplanes we often take for granted the fact that we can fly but there's been no improvements of in aviation anywhere close to the improvements we've gotten in computing power I like to ask well what would the speed of planes be if they doubled every two years over the past half century like computing power is done well they'd be at six times the speed of light it's physically impossible for Aviation to produce that level of improvements but the chip industry has done it and if you look across all other types of Technology all other facets of the economy nothing else has come close to the rate of improvements driven by Moore's Law which is why the capabilities that Moore's law has made possible have been applied to every other facet of the economy the chip industry has been colonizing the rest of the economy taking over the world because everyone needs the capabilities that it alone can provide Computing sensing Communications which has transformed basically every segment of the economy in basically every country and so when I argue to you that Intel is is perhaps the most important company that's existed I think you'll begin to understand why I think that because not every company they can claim to have influenced Intel but Intel can plausibly claim to have played a central role in driving forward the technological advances that today every single person and every single company relies on I think one of the great failings of the chip industry over the past several decades has been to not realize just how Wild magical and cool it is uh to undertake the type of accomplishments of the chip industry takes people in your position are just accustomed to taking for granted exponential growth rates without thinking about how extraordinary it is in the context of the rest of our lives before I began studying the semiconductor industry I'd only seen exponential growth raids in math class which is the only other part of human existence that has a regular exponential growth curves and yet the chip industry delivers them on a regular basis it's so second nature to people on this uh teams call that you don't even think about how extraordinary it is and so what I would first like to encourage you all to do is to take a step back and just wonder at the extraordinary accomplishments that the semiconductor industry and Intel specifically have helped produce over the past half century because I think you can't understand any of the key developments in recent history without putting the development of computing capabilities absolutely at the center of that analysis and that's true not only when it comes to things we think of when we think of technology smartphones PCS data centers the Internet it's also true when we think about bigger social forces which themselves have been transformed by technology that semiconductors have enabled one of the reasons why I decided to write Chip War was when I began to realize that semiconductors weren't only important for my PC or for my smartphone but they were important to understand all of the key Global trends that were unfolding on a daily basis in in newspaper headlines all around the world for example when I learned that today China spends as much money each year importing semiconductors as it spends importing oil I realize that you can't understand the structure of globalization without putting semiconductors at the center and I think that single data point puts very clearly into um into context Pat's argument about semiconductors being as important as oil well in terms of trade data that's just a fact so my conductors are just as important as important as oil for uh several of the world's largest economies and so if you look at big shifts in the world economy semiconductors have been a fundamental driver uh the success and the failure of not just companies but also entire National economies and indeed when you look at the ways that globalization is changing again here to semiconductors are at the center second I came to realize the way that semiconductors have played a central role in the production of military power as Carl mentioned there's been a deep interrelationship between the development of computing capabilities and National Security implications and indeed if you go back to the early history of Intel indeed the pre-history of Intel when Bob Noyes and Gordon Moore were at Fairchild semiconductor which they'd founded before going on to found Intel there was a very deep and intimate relationship between that company and the early Space Race and the early Cold War National Security State indeed it wasn't a coincidence I don't think that that Fairchild got off the ground and no small part by selling semiconductors that went into guidance computers that were used in space and missile systems and although today companies in the semiconductor industry generally Focus most of their sales on civilian products like computers like data centers like smartphones it's still the case that a key demand for semiconductors comes from military and security applications and having access to the most advanced Computing capabilities the most advanced sensing and Communications capabilities is absolutely fundamental for defense and intelligence agencies especially as they increasingly try to Grapple with use cases of artificial intelligence so just like in the early days of Intel's co-founders I think today as well when you talk to officials in the defense and intelligence World they see semiconductors as not just important for civilian Technologies but also is critical for National Security implications and indeed as as Pat and Carl both alluded to today the entire world is beginning to reshape the way that it produces semiconductors is nearly every country and nearly every government have become concerned with questions of supply chain resilience and of the geopolitical balance uh that semiconductors play a role in that'll come as no surprise to anyone this team's call that the manufacturing of semiconductors has got an extraordinary concentrated in a small number of companies and in a small number of countries and when you look across the world as the US Europe Japan Korea India and many other governments are trying to engage with the semiconductor industry a key Focus that you see is diversification and supply chain resilience it's being two Central themes of government policy and governments are interested in these two topics because they realize that semiconductors are as important as they've ever been not only for their security notifications but also for their broader economic ramifications and I think as a final point to wrap up on the fact that semiconductors are being applied to a ever broader sphere of devices makes them ever more Central today in comparison to several decades ago the pandemic era chip shortages I think helps people understand the extent to which today almost every manufactured good has at least one and often dozens or hundreds of chips inside most people even I think car company Executives were astounded to learn during the pandemic that a new car will often have a thousand semiconductors inside and that points to the extent to which the Computing capabilities that chips enable have become pervasive in every device there are increasingly greater numbers of chips doing many different things some very expensive some very simple but enabling not all of the capabilities we rely on I like to think about just the number of semiconductors I interact with over the course of my daily life and my alarm clock and my refrigerator in my car and my microwave even before I've picked up my smartphone there are dozens hundreds probably even thousands of chips that I've touched to say nothing about the chips that are in the telecoms infrastructure and the data center is that increasingly store and process all my data and that's only possible it's only possible to have chips in all of these different devices because the cost to compute has been relentlessly driven downwards and it's been relentlessly driven downwards because of Moore's Law and so when I say that Intel which has been the driving force behind Moore's law for half a century is the most important company to understand the history of the last 50 years that's exactly what I mean because the central Trend in uh in the development of computing which is Moore's law more powerful more cost-effective compute has been important far beyond this one company transformative though it's been for Intel it's been important for everything because all the devices that I use that rely on computing power have benefited from the extraordinary advances in semiconductor design and Manufacturing that have enabled Moore's Law and so I think when the semiconductor industry and when Intel looks at the role that it plays in the world I think the entire chip industry can do more to recognize it's absolutely Central importance I think Pat's right when he says that chips are the oil of the 21st century it's it's actually more true than I I think that statement um gives credit to because oil is useful for certain parts of the economy for cars for example uh for providing energy but chips aren't almost everything and so I think uh I think it's certainly right to say that uh chips will be the critical technology of the 21st century I think they already are the critical technology of the 20th century and recognizing and focusing on and streamlining and diversifying the ways that they're produced and the places they're produced I think will remain a central focus of both industry and of government policy makers for many years to come so thank you Carl for uh for the invitation to join you and I'm looking forward to the discussion today great well I thank you for painting a a wonderful Mosaic picture of the importance of our products our company and the industry to the world and you know the last 50 years has been phenomenal and you think about what we have today I can't imagine or couldn't wait for the next 50 years there's going to be some really incredible stuff that the people at Intel are going to be front and center on delivering so thanks Chris and uh we'll we'll move on here to uh the question so that I have for our panel members and we'll be back to you in a couple of minutes with a question uh the winners in this competitive race will be companies who are most effective in building filling running and repeating state-of-the-art semiconductor factories yes that's right repeat because there will be another generation of advanced semiconductor manufacturing after euv this promises to be a competitive cycle Without End and my first question goes to Al Thompson hell I've always been passionate about hearing from our people who face the customer every day you are in a unique position you see U.S policy makers as your customer every day and Bill means it takes Capital to play in this competition the White House recently announced 460 companies have expressed interest in the 50 billion dollar chip Act that's good news but if they each need a billion dollars we are short funding by an order of magnitude will the U.S government be competitive with the Chinese government I'm curious will Congress invest in state-of-the-art semiconductor wafer processing round for round with no strings attached so one it's it's great to be here Carl I you know it's it's very interesting kind of looking at this chips act Journey from from my vantage point and particularly the amount of advocacy that that patents do and others have done to put us in the position I'll answer that question a couple of ways right I think one the reason why the U.S government
basically passed the chips act and then the chips and science act posed covet was a recognition of as Chris said the value um and the critical role that semiconductors play for both economic and National Security and I think they realize that it is a global competition they were very concerned about the over concentration of manufacturing in Asia I would say it's broader than just China and I think they decided we need to come with a whole of government approach to basically incentivize and drive more manufacturing here to um the research apparatus of not only the the Commerce department but the Department of Defense to uh look at ways to continue to innovate and then I think finally to look at ways to leverage the supply chain because our our supply chain is very Global to ensure that we are partnering with key allies across the world to ensure that we um create more supply chain resiliency and balance so I think we're going to be very competitive I will say one of the things that ships did spur that Chris mentioned that we highlighted a lot when we were pushing for the passage of the chips and science Act is this is now a global competition for incentives right it's not just India it's Japan South Korea who in Taiwan You could argue had generous policies already in place have increased what they're already doing um you know the EU is now implementing their chips act and so I think like it or not we are in a race to ensure you can secure um end-to-end supply chain resiliency and Manufacturing string of semiconductors on your Shores and that's a good thing I think as it relates to the 460 companies and the 200 billion dollars of investment that the chips science access for I you know I like to say I'll take Intel's hand any day of the week um you know I think you know Chris highlighted the importance of Intel but I can tell you and I think it was you know I know Pat um shares some stories when he was at the State of the Union but to hear the president of the United States essentially say Intel is an essential company um I think says says a lot we we take our responsibility in trying to work to secure our funds and win this competition pretty seriously but again I think we have um a unique role a set of and a set of assets that nobody else in the industry has and I think we're going to leverage those hopefully get the funding that we need um I will say you know there are there's never Government funding without strings attached that's just that that doesn't happen um but the nice thing about the funding and the chips and science act are the strings that they're attaching are all the Intel's benefit right Intel is the one company I believe that's applying for chip funding that can essentially achieve all the different objectives that secretary Armando and the president have laid out for the program so our strings that they're attaching in terms of Workforce Development diversity and inclusion environmental impact all work in our favor so this is probably one of the few times where are there strings yes but the ones that they're attaching we pretty much do on a daily basis as a company and I think it's going to cure to our benefit you know one of the things Al that uh Chris pointed out in his book was the development of the semiconductor equipment industry that supplies uh our needs for factories and so forth being a diverse worldwide supply chain you know we have important companies in Taiwan Japan the Netherlands and so forth it kind of raises a kind of a political uh aspect of that and I just wonder if the U.S government will actually see the potential of a worldwide deployment of this supply chain for Semiconductor equipment as a path for a possible to taunt as tensions build among all the various uh Bad actors if you will in the world and the political tensions that uh we we all hear about I mean I think one there's a there's a strong recognition I think that the US government has learned probably in the last 15 18 months about the global nature of the semiconductor supply chain and frankly the the significant role that both the Japanese and the Dutch play um I'll tell one quick story just to kind of show how Intel's played a role in informing that um one of Pat's earlier meetings with when in his tenure was with the National Security advisor and he basically walked him through the role that Japan and both um the Netherlands play uh in the ecosystem and made it clear that as you're looking at your policies right things have to be multilateral because of the role that they play and I think you know fortunately the the US government has taken that to heart I think they've been highly engaged with both the Dutch and the Japanese government on how you manage export controls of China for example which I think will be beneficial but I go back to again one of the most under sold parts of the chips in science Act is the multilateral Security fund and what that really is is the U.S state Department is going to have about 500 million dollars to spend over the next five years and I think there's going to be opportunities to leverage the supply chain the existing Partnerships that can exist diplomatically to um to essentially I think try to troubleshoot and get ahead of future problems right that's one way I think the other thing you see is that the uh the EU and the United States have What's called the trade and Technology Council and while that covers a whole range of issues semiconductors has played a role in particularly I think one of the things that we're pushing for is greater coordination as the EU and the U.S implement the chips act so that's a long way of saying that I think you know that cooperation is already happening and I think there's a clear recognition now in the U.S government as they look at
things geopolitically that we have to partner with our allies and particularly Japan and the Netherlands as it relates to equipment in the supply chain in general good so you know I read a lot about uh you know continuing discussion of amendments and changes to the chip Act what is really going on in Washington today regarding the chip Act um what what cabinet level member is responsible for driving the chip act uh who do we go to as the uh you know top dog here that we can have a session with and get something done multiple no I I mean I think the as we look at it the chips Act is really kind of a three agency uh built I mean the Commerce department number one secretary Gina raimondo Who uh her team is basically implementing the law but also reviewing all of the applications so uh as Stu well knows and and Jim we spend a lot of time with the chips program office who reports directly to her um she is a clear big deal believe it or not the Department of Defense plays actually a really big role in ships um first there's r d money set aside to basically start to drive Innovation hubs around the country to support DOD initiatives but the other thing that doesn't get talked about a lot that um students very involved in is in the original chips act there's actually legislative policy designed to better align um the department of defenses long-term semiconductor needs because one of the challenges that has presented itself um is the fact that as Weapons get more smarter and you need more chips it's very expensive to manufacture those for the Department of Defense in a secure way and so they have to leverage the commercial manufacturing process I mean I always kind of chocolate was one of the best parts of the book that I thought was the link of Bill Perry who became a secretary of defense under Bill Clinton but cut his teeth as the under Secretary of Defense of research engineering I can tell you the Intel team will advice spends a lot of time with a OSD already and so I would say DOD is number two because of the role that they're going to play and then again it goes back to State But to answer your question we spend a lot of time with the Commerce Department um because they oversee I think almost all of the funding for the bill and I think look something's they are moving things are going to get done I know Pat has shared that you know we have essentially are on track to submit applications for what we're doing in Arizona Ohio New Mexico and Oregon um I mean I can promise you my colleague you know Rex later peruse my boss and others in the government Affairs team we spend a lot of time moving to answer their questions to hopefully get to a decision point so uh that's one part I think the other thing is you will start to see Congress focus on oversight of the program um you know I think the law has passed it's been signed the Departments have been I'm sure keeping I know keeping the hill abreast on kind of what's going on but I think the thing you'll also will see is greater oversight as the funding starts to go out the door and that's something that we're trying to you know I think shape in a way that ensures that the issues that we want to be addressed get addressed but you don't want to slow down or enter the program just given the global Dynamic and the things that are going on in terms of the need to get more capacity online great it's good to hear that Commerce and defense has their hand on the tiller so hopefully that'll work out well for us so I'm going to move on to Stu now for some questions um you know Stu Phil means to me frankly Phil this looks like a sales opportunity in terms of the number of factor factories and the amount of spending you guys are committing to the factories you know empty or underutilized uh Fab means losing to the competition the x86 architecture is a strategic Advantage for Intel because you have a demand for a large number of wafers to service x86 and x86 computers will be the bulldozers and the cranes that build out software for the two major markets that are developing in parallel right now that everybody's excited about Ai and VR I want to understand if intel has the will to not say no to any new consumer or infrastructure projects that the market may present to Foundry services and commit wafer capacity to it well thanks for the question Carl I think as many of the Intel Alumni network knows I was raised on the Carl Everett School of Management uh in terms of sales of commitment so when you make a commitment you honor a commitment and you follow through on it and I think for us what we're doing in our best is we're pursuing a number of opportunities but there's one area I'd like to just maybe expand a little bit in terms of combat x86 our focus in IFS is in more ways more focused on arm than it is on x86 because today 80 of the Wafers that run through tsmc's factories have an arm micro processor in that and this is why you know just a few months ago you saw us announce a major partnership with ARP which is something that in my first Duty as you know as I grew up with you and all the other people in the Alumni network the fact that we would Embrace harm as a company was just anathema and now pragmatically to your point about filling factories if we want to fill factories if we want to compete with tsmc we must have a viable partnership with our you know they're shipping 9 billion units a quarter right now of licenses so we will play with them and when we commit to a customer especially on The Foundry side it's a committed capacity Corridor because in order to be successful and found your business you know it's first and foremost a trust business if our customers don't trust to put their interests even ahead of our own business they're going to be want they'll be reluctant to partner with us so as we develop these relationships we're going to these customers saying we will commit a quarter of capacity to you that's in violent it's yours we're not going to go take away so the reason you see you know my supervisor going out and spending all the money he's spending is we want to make sure we have enough capacity in place they have the shells ready I think many of the folks on the call remember Jerry Parker's laws always be one shell ahead of the demand this time we're going to be multiple shows ahead of the demand because we're going after some very significant volumes and we'll work with the major players such that when we commit to something we will honor it now luckily you know we're coming from you know from a sales guy point with lots of opportunity because you know we're not a major presence just yet but we look to you know absolutely fill all those factories up you know over the coming years and our goal is to be the number two Foundry by the end of the decade and we have a road map laid out of processes and products and packaging and capacity to go do that we know which customers buy you know at that level and we're courting them as hard as we possibly can so I feel pretty good about our chances you know the customer base wants an alternative to tsmc as Chris alluded to earlier there's a need for supply chain resiliency in the world there's a need to rebalance where these kind of processes are running and we offer that opportunity our cut to our customers we also offer to offer the opportunity to negotiate right there's not much negotiation going on between tsmcs and customers right now when they have 90 Market segment share uh you know they will you know enforce their rules on this industry and so the players in the industry want Alternatives want resiliency and I think as we get to five nodes in four years which we're absolutely marching toward and have process superiority you know in the coming next few years we'll have another reason for the customers to come to us so I like our chances and we will not take you know we're not turning down orders arm x86 space it doesn't matter to us what matters is you know the quality of the commitment and the you know quality of the customer working with us to go bring that volume forward because we can't take Bets with customers unless they're going to be successful too so we'll be careful we'll be judicious but we're also going to embrace you know every major opportunity that we can yeah I think uh you're you're right on there the uh the trust has to be rebuilt and the new business model has to be uh established okay and you're going to need all the help you can get there from everybody which brings me to kind of my next question uh our friends at the government um do you think you can work with governments to build creative incentives to um you know having to have customers that haven't bought from us before okay uh take risk on Intel state of the art Wafers you know for example a nominal uh tax incentive for any customer who buys and accepts delivery of any product based on Intel state of the art Wafers would be a great signal from our friends in Congress okay uh man it's kind of like a potential for a Tesla moment that you certainly got behind electric vehicles why wouldn't they get behind semiconductors at the same level of waiting okay uh that they're behind electric vehicles if it's a National Defense priority you know I would love for that to happen and I I think in uh you know our late evening conversations that I had with Al and the other members of our DC team we talk about that but the reality is we have to deliver on chips at Grants and money now we have to show that it actually works as Al mentioned there's a lot there's going to be a lot of oversight in this program and there's also the usual you know Republican Democrat kind of debate back and forth as to what's better public policy so there's a lot of eyes on this industry right now to actually show that this can work and if it works and the money gets Grand we start building factories I would hope that would be a logical Next Step because as you and I both know constraints come and go this industry and you know we're incredibly concerned about constraints coming forward in the next three or four years to potentially do you know World economic situations we need to have resilient Supply chains and when things get Freer again people sort of forget about that and so any help in public policy to keep that moving forward even though you can the tensions need a supply chain have somewhat dissipated would be helpful but you know job one is get the grant money out now get them apply to our factories get our factories up and running and then see where we can go from there okay great let me move on to uh Jim Evers uh uh for a couple of questions here uh Jim run means setting the speech which the industry competes how do you make Jerry Parker's elephant dance after 50 years of proprietary wafer processing you're now moving to a new open wafer processing models Stu's just describing you still have a big order for x86 wafer starts that's good news the next challenge is is responding to you know many new Foundry customers and small lot wafer runs you have a development underway and tools and process that handle the many diverse products and radical changes that will be required great thank you thank you Carl for that one okay before I answer that question I just have a couple things I want to say first Chris it's great to be with you here I I got your book about a year ago from my um mother-in-law and the page two of the cast of characters uh Andy Grove was on there and I read through it like a spy novel and it was a go-to Christmas gift last year so very very relevant and for the Intel alumni thanks for having me back I did a session with you last uh summer so good to be back with you we absolutely have development underweight for the scalene piece of this one for the multiple products that we had as we were moving into Foundry space I've hired new Talent out of The Foundry companies that are existing I was asking about the number of products that they run and in Arizona where I was a factory manager prior to Ohio we had we had had similar number of products that were there because we were running three Technologies in parallel um uh and so each of those had a product Suite associate with them does it get even bigger as we mature the Technologies and you add more customers it does but I I think the technology piece is the easiest of this those that remember 200 millimeter where we used to load Lots by hand know that we went to an automated Material Handling system so now everything is kind of controlled for the reticles uh it'll be the same thing that'll also be manually controlled in an age we talk about AI you know artificial intelligence we are not only the manufacturers of the chips that go into artificial intelligence but we have the automation that controls our manufacturing as artificial intelligence what I mean by that is our manufacturing execution system we'll think through where every single lot needs to be in the factory and and how to get it to the next piece of equipment and how to land that reticle where it you know aligned with that particular lot and and run through that one that's probably something that from a roadmap standpoint we're really good on if we think about that scaling though of the number of customers that we have in my mind the bigger challenge is I've gone through the learning of being you know a manufacturer and Foundry I think uh one of my big learnings in the last few years is that we are a customer service organization and so how do you treat all those customers like they are the most special only customer in the world on the planet and to be able to scale that way for the interface to get them the data that they want to get them their customizations they want I think that's a even bigger challenge for us that we are facing or addressing right now you know Intel has announced this SketchUp strategy of five nodes in four years I can imagine a large increase in x86 product roadmap line items more nodes faster would seem to drive more products tell us about your five and four game plans are the factory and the product design teams working together to simplify the roadmap does a proliferation of product line items help you achieve world-class node performances you were just discussing yeah so this is really good I'm happy to report you know we talk about five nodes in four years that we are on track to do that we had to do five nodes in four years to catch up now let me put things in terms of the you know the terms that the alumni would be around for to kind of do the math on the five nodes in four years because what's old is New Again in terms of we talking Tick Tock for those who've been here a long time remember tick was you know brand new major architecture talk was you know a brand new major process node but you would do derivatives in between and it was every two years so if you look at the five nodes that are existing that they're broken up into uh to the separate nodes in the middle there intel four and until three would it be a major major process node all new Tooling in a different set of factories and then you would customize that again in Intel three same thing as you go from Intel 28 to Intel 18a so you can then take that and split that up and have different factories that are running into 18828 combo into until 4 Intel three combo and of course we have our Intel 7 that's running in multiple factories today so that dividing up of the work on the factory side is pretty easy to do there are some parts of our business that have to kind of service all of that mention the product groups as well as let's say the mask operations so a couple years ago we doubled our mask operation capacity by opening another site in Oregon so um yes it's a challenge for us but it's necessary to do and we're executing well on it super help me understand how much investing investment funding by Intel and the industry will be required to develop this geographically diverse semiconductor equipment supply chain what are we going to do to make sure that Intel has the people Talent to run the Fabs to develop the technology required for the new nodes and move them into production consecutively here I think people could be the next big story here yeah that's a great question if you look at the market growing to 1 trillion if you look at the fact that across all of the different even outside of Intel big project announcements whether it's Intel in Ohio and Arizona there's so much growth that's going to have to happen investment that it does come down to people and if you look at some of the latest semi reports that are out there it says that we're going to grow 115 000 jobs by the end of the decade now here's the scary part that same report actually said that 76 000 of those could be potentially unfilled so Intel has taken an approach that hey Talent can slow us down even as quickly as building space can let's invest in an example in Ohio we committed 50 million dollars into the education system and we've already put 17.7 million dollars out there for you know 80 different institutions across Ohio because we need to develop and cultivate that uh the talent pipeline so we're working on that now fantastic Chris I got a question for you um I in Reading chip Wars you cite two examples of governments trying to create homegrown Industries to compete in semiconductors one in China One in Russia both failed Industries built by government governments often failed because of government the China and Russia Industries happened while the U.S government was much less involved in the semiconductor industry in the last 40 years than they were in the first 10 years in the early Minutemen days how do we protect our government from failing in the next era when they seem to be signaling a desire to be more Hands-On in funding with an enormous increase in complexity here well thanks for the question I think it's a critical a critical challenge that the government faces because I'm on the one hand there's no doubt that government has repeatedly played a role in defining the direction the chip industry goes and that's true from the miniman days it's true from the invention of finfet transistors there were government involvement at every step in that process but it's also true that that its companies not governments that make semiconductor manufacturing profitable uh it's it's also true that it's companies not governments that understand where demand is and how to meet it uh profitably uh and and efficiently and so I think governments need to be very very careful in the role that they try to play I if you look at China right now which is investing as much money as ship industry from government funds as the rest of the world combined I think it's roughly one ships act per year of funding that goes into the Chinese chip industry the chips act in the US will play out over half a decade what you find is that although the impact of that massive sum of money is real uh the the return on investment of every dollar that China spent is very bad I think the Chinese state has been one of the worst Venture Capital firms in the history of the entire technology industry and I don't think the US should try to repeat that um however I think it's also true that until recently other government were being quite generous with their semi Hydro Industries in the way the US was not and indeed there's an extraordinary study that was assembled in 2019 by the organization for economic cooperation development the oecd which measured the amount of government funds that different chip firms from different countries receives and one of its findings was that there were U.S
companies that received more money in manufacturing subsidies from other governments than the United States and so in that context it's not hard to understand why many companies made decisions to build factories in other countries and so I think if you look at the chips act today at the highest level with the US government is trying to do is is to level a playing field that right now is very unlevel and I think that's a role that we can feel relatively comfortable for governments playing even though I completely agree with the spirit of the question that we should be very cautious about having governments play any more of a Hands-On role thanks and at this point I'm going to switch to questions from the audience okay and Al you're up first I have a question from Omar Sadat companies or industrial sectors that are closely linked to Government funding defense narrow space for example are often inefficient since they are forced to optimize solely for for securing government funds will the impact be on Intel's long what will the impact be on Intel's long-term competitiveness as we embark on a strategy of increasing government engagement I believe positive to kind of build off of what what Chris just said I think there's another Dynamic that um is is important here right and if you look at as we were talking about Moore's Law And basically now that the three companies that can make a Leading Edge chip you're down to also three countries right it's it's the United States it's it's Taiwan and it's Korea and so I think the way that ships Act was structured where you're really talking about leveraging a massive investment tax credit and also specific grants to incentivize is the right balance that allows us to basically maintain our commerciality without basically going through what I would you know coming from a defense company some of the laborious processes it takes to basically buy things in the in the defense world and so I don't think you know I think if we're going to rebalance a supply chain a public-private partnership with the government's going to be required and I think we've struck the right balance at doing that in multi Parks across the world next Stu we have a question from cliff perkhouser for us alumni who were at Intel during Andy Grove era a common question is why did Intel lose its leadership Stratus in the 21st century was Andy just an exceptional leader and his successors were not able to follow I think it's a great question as I think some of the folks on the call note today i'm back for my second tour Duty I left Intel in 2014 to go run manufacturing HP for a number of years then came back to help Pat rebuild the processes and the approaches we used to make the company successful back during the Grove era I think there are a couple things that in reflection on this question which I also get asked a lot we made some fundamentally bad choices and only takes a few really bad strategic choices to alter the Strategic Destiny of the company one of the things that Chris talks about in his book is although we were a Pioneer in the UV we had the first investment in asml we decided to go a different direction may you remember the debate over 450 millimeter Wafers versus 300 millimeter wafers and I remember the meeting where that decision was made and just um we lost out an opportunity and I think uh that combined with a lack of what really made us successful in terms of focus Chris talks about his book with tsmc they're singular focused on cost and Moore's law in many ways is a law about economics and as you know we look at where we're at today we have to strive to get cost out of these products in order to make doubling the number of transistors on affordable economic option so a couple of bad decisions one that Paul or Delaney admitted to you know publicly to the Alumni network was a decision not to do the Apple iPhone based on the fact that we didn't think we could be cost competitive and two was the decision not to go after euv very heavily and then third decision was getting to businesses that we didn't frankly have a place to be in drones clothes yeah they're not Central to changing the world and I think with Pat's leadership and Pat's leadership and Sal's the reason I came back is we have the focus now we have a strategy that honors what we had you know in terms of our existing PC and server business but doesn't keep us bound to it and that's why we're out pursuing boundary business today with the ferocity that we're doing because the strategy of undergrowth would not work in today's world not with tsmc and you know the fundamental process advantages it has and that we hope to catch up on So Different World different strategy I know if Andy would have been able to quite manage his way through that imagine the conversation Carl we had had with Andy and said hey look Andy we want to do a different architecture today we think that's the right thing to go do and we're going to open up all of our factories I I don't know if any would have been able to you know do well with that conversation I'm sure we have you know a pretty Groban debate about it but I think now Pat's face the fundamental realities of where we're at his company he's you know you know promoting this idm2 strategy that we all I think fundamentally believe in and it's starting to you know bear the fruit as as we get more and more confidence in our approaches with Foundry customers I think you'll see you'll see this play out in a dramatic new way for the company and its customers its employees its stockholders okay Jim I'm going to come back to you in just a second because Chris has a flight to catch and he'll have to leave here at five o'clock Eastern I believe so Chris I've got one final question from the audience uh for you to uh field uh today um it's from Milt Walker he says after reading Chip War and enjoyable and insightful read it was hard to miss the Fingerprints of U.S industrial policy on the storyline it seems we may be at a similar inflection point with respect to the industry do you see an important differing factors from the past that might drive different strategies from Andy Groves trade-off between RAM and CPU one of the I spent a lot of time trying to understand Andy Grove's thinking in the 1980s when that trade-off was was being made and I think one of the the things that was really striking was the extent to which he talked about the decision to Pivot away from memory as an emotionally wrenching decision that he'd helped build the company around producing memory chips one of his executive team members described the decision to Pivot away from memory is like Ford deciding not to manufacture cars it was that much of a cultural shift or seen that way to Grove and and many people around him I'd be interested to hear that others in this call uh would would Echo those sentiments but what I think he fundamentally got right was that there was in the memory business a pathway to continue a competition that was not really wittable I you could keep pouring money into the memory business but the competition was both tough and headed in a direction that was not going to be profitable for the company there were too many players in the memory business there was no way to differentiate a Competitive Edge and instead the Brilliance of the strategy was to focus on a type of semiconductor where there was far less competition and far more scope for growth and that was the microprocessor and CPU um and so I think that has actually real lessons for today because I'm on the one hand the industry in town particular but the entire industry is focused on the competitions that we're in right now and those are important competitions for every company to want to win but actually the thing that Andy Grove did really well was to understand when to leave a market that wasn't going to be profitable and pivot to something entirely different even when it felt like it was in some ways a rejection of the heritage of the Intel that he'd helped build over the course of the 1970s and early 19 80s and so reading uh the the letters and the memos that he wrote at the time and his interviews from that time you you get a sense of the the emotion that was involved in that shift but fundamentally it was the right shift to make even though it involved such massive changes that until at the time and I think for every time I entered a company but uh perhaps also for Intel right now there's there's Echoes of that of big changes in the way of doing business that have to be undertaken that are hard controversial emotionally uh uh uh as well as institutionally but nevertheless that's uh that's how technology companies have to Pivot when technology begins to change yeah well said well said and we appreciate your time today and spending your time with Intel employees and alumni we look forward to having you back again sometime and we look forward to the next book too that Pat mentioned I can't wait to get a copy well thank you it was an honor to join you today our pleasure Jim I'm going to come back to you we have a question from Chandra berance Baranowski pardon me for the mispronunciation will newer manufacturing equipment scanners Etc ever become so expensive that Intel or that that the total available Market of semiconductors can't stand multiple profitable foundries I love this question because uh before as a factory manager I was a litho department manager and lithography is staggering the fact that you could have a piece of tooling now reaching around a quarter B is like unbelievable and so let me answer it in two different ways one is that investment that's required absolutely has an impact on companies we've had consolidation in the industry because of the fact that the investment that's required you know today only three companies in logic can do that right gsmc Samsung and Intel and now what I do think though is that there will always be a need for two companies because customers are going to want an option but there's no entitlement for the third so we have to be hungry to be in that top tube and reach that by the end of the decade now the question asked about the total available we ever get to the point where it's so expensive you can't do that what there I'm going to Pivot to really it's it's more of a question on do you believe in Moore's law that
2023-09-18