Good day, ladies and gentlemen, and Welcome to the Amkor Technology Q1 2025 Earnings Conference Call. My name is Diego, and I will be your conference facilitator today. At this time, all participants are in a listen-only mode. After the speaker's remarks, we will conduct a Q&A session. As a reminder, this conference is being recorded. I would now like to turn the call over to Jennifer Jue, Head of Investor Relations. Ms. Jue, please go ahead. Thank you, Operator. Good afternoon, everyone, and thank you for joining us for Amkor's Q1 2025 Earnings Conference Call. Joining me today are Giel Rutten, our Chief Executive Officer, and Megan Faust, our Chief Financial Officer. Our earnings press release was filed with the SEC this afternoon and is available on the Investor Relations page of our website, along with the presentation slides that accompany today's call. During this presentation, we will use non-GAAP financial measures, and you can find the reconciliation to the US GAAP equivalent on our website. We will make forward-looking statements about our expectations for Amkor's future performance based on the environment as we currently see it. Of course, actual results could differ. Please refer to our press release and SEC filings for information on risk factors, uncertainties, and exceptions that could cause actual results to differ materially from these expectations. Please note that the financial results discussed today are preliminary, and final data will be included in our Form 10-Q. I will now turn the call over to Giel. Thank you, Jennifer. Good afternoon, everyone, and thank you for joining the call today. Amkor delivered revenue of $1.32 billion in the Q1, at the upper end of our guidance. Communications revenue exceeded our expectations, while other end markets performed as anticipated. EPS was $0.09, impacted by higher R&D costs, including accelerated development in RDL technology for additional programs scheduled to ramp this year. We are closely monitoring the tariffs and trade regulations. On tariffs, our global manufacturing operations are largely unaffected at this time because the majority of our facilities operate in free trade zones, and we only ship a small amount of semiconductors into the U.S. The main uncertainties revolve around potential disruptions to our customers' supply chains and potential demand swings for consumer-driven end products. Our diversified global footprint and long-standing partnerships allow us to help customers work through this complexity. On trade, we continue to monitor evolving U.S. export controls, including expanding technology restrictions. As an approved OSAT by the U.S. Department of Commerce, we are an ideal partner for our customers. The Amkor team is handling these market dynamics with agility and is focused on executing our long-term strategy. Our strategic framework is based on three pillars: strengthening our technology leadership, expanding our geographic footprint, and partnering with lead customers in growth markets. This strategy helps us to leverage our competitive advantages that differentiate Amkor as a Tier 1 OSAT. As a technology leader in advanced packaging and test, we are a trusted partner for the largest semiconductor companies to enable their product roadmaps. Our leadership and engineering expertise allow us to address manufacturing complexities through co-development with customers, utilizing our broad technology portfolio. High-performance computing and AI are driving technology innovations and the need for advanced packaging. In addition to our engagements in 2.5D and RDL technology, we are collaborating with our data center and networking customers on co-packaged optics and photonics solutions. Amkor is engaged in photonics in pluggable devices and in next-generation advanced c o-packaged optics devices with heterogeneous integration of optical engine chiplet dies in a module format, improving power, bandwidth, signal, and space efficiency. The growing complexity of devices requires a comprehensive testing strategy. This quarter, we confirmed plans for a turnkey test solutions expansion on our K5 campus in Incheon, Korea, where also our main R&D center is located. We expect the first phase of the expansion to be operational in the existing K5 facility by the end of 2025. The next phase, including a new building, is expected to be operational in the H1 of 2027. Our second pillar aims to expand our geographic footprint to align with customer and industry supply chains. To support accelerated demand for advanced packaging services in the US, we are evaluating options to increase scale and expand our technology offerings. We continue to broaden our partnerships with lead customers and silicon foundries, and we are on track to begin construction of our Arizona facility in the H2 of 2025. Beyond these significant projects, we continuously evaluate our footprint to ensure we are optimally scaled with the right technologies in the right locations. In our third pillar, we are strengthening collaboration with leading semiconductor companies to facilitate early engagements in product development and to co-develop innovative advanced packaging solutions to enable fast time-to-market for lead applications in the industry. In communications, we have primarily served the premium-tier smartphone market for both the iOS and the Android ecosystems. With AI applications shifting to edge devices, we expect innovation within smartphones will accelerate in premium-tier first, notably in apps processors and connectivity applications, both requiring advanced packaging. Within the computing end market, we support customers across the data center and PC landscape with advanced packaging and test for multiple devices in AI and high-performance computing applications. The accelerated transition to a new AI GPU product family and the expanded trade restrictions tempered our outlook for this year. Demand for prior-generation devices will continue for the year, but volumes are difficult to predict due to the impact of export controls. Overall, we have a robust project pipeline in our computing end market, including our new 2.5D switching customer that continues to ramp volume and our first AI CPU programs ramping on RDL interposing technology, which gives us further confidence in our long-term outlook. While the automotive and industrial end markets are still recovering from weak end market demand and elevated inventory levels, the demand for advanced packaging solutions continues to be robust. The main driver is proliferation of ADAS and infotainment functionality across the car ranges. Amkor is a leading automotive OSAT provider with a strong track record in automotive manufacturing and advanced packaging technology, where we have a solid pipeline with multiple customers for new radar and LiDAR applications, ADAS, and CMOS image sensor programs. In the consumer market, long-term drivers include the growing demand for wearables and connected devices. Our advanced SiP expertise positions us well for consumer growth, and between our established Korea facility and new Vietnam facility, we have alternative locations to offer customers this technology. In summary, we remain confident in our strategy to achieve long-term profitable growth. We are actively supporting customers to resolve supply chain challenges in the current dynamic environment, and we are focused on executing our strategy. With that, I will now turn the call over to Megan to provide more details on our Q1 performance and near-term outlook. Thank you, Giel, and good afternoon, everyone. As Giel mentioned, we are navigating a dynamic environment. The team is focused on adapting to change and managing elements within our control while continuing to execute on our long-term strategy. Amkor delivered Q1 revenue of $1.32 billion, reflecting a 3% year-on-year decline. Revenue in our communications end market decreased 19% year-on-year, primarily driven by lower revenue within the iOS ecosystem. A new SiP socket we are co-developing for the next generation of smartphones is on track to begin production by the end of June. For the Q2, communications revenue is expected to increase sequentially, reflecting efforts to optimize line utilization. Revenue in our computing end market increased 21% year-on-year, driven by multiple engagements across data center, networking, and PC customers, as well as accommodating dynamic build patterns for AI GPUs using 2.5D technology. Computing is expected to grow sequentially in the Q2, driven by strong demand for new PC devices. Revenue in the automotive and industrial end market declined 6% year-on-year but remained stable sequentially across both advanced and mainstream products. Our project pipeline with our largest customers in ADAS, infotainment, and other advanced applications is robust, and we have begun ramping solutions that we expect to drive growth for this end market in Q2. Revenue in our consumer end market increased 23% year-on-year, driven by a continuation of the hearable program utilizing advanced SIP technology that launched in the H2 of last year. For Q2, we expect the consumer market to be relatively flat sequentially. Q1 gross profit was $158 million, and gross margin was 11.9%. Gross margin was lower sequentially and year-over-year due to the impact from lower volumes, with resulting factory utilization in the low 50s. Operating expenses came in higher than expected at $126 million, primarily due to the increase in R&D, which Giel mentioned is associated with development of new technology, including RDL, that is expected to ramp later this year. Operating income was $32 million, or 2.4% of sales. Cost containment measures are in place globally, and we have continued to be profitable throughout this semiconductor cycle. Net income was $21 million, and EPS was $0.09. Q1 EBITDA was $197 million, and EBITDA margin was 14.9%. Amkor has exhibited strict financial discipline and maintains a strong balance sheet. We ended the quarter with $1.56 billion of cash and short-term investments and total liquidity of $2.2 billion. Our total debt as of the end of the quarter is $1.15 billion, and our debt-to-EBITDA ratio is 1.1 times. The strong and flexible balance sheet we have built enables us to enhance shareholder returns by investing in organic growth, including technology development with lead customers, exploring strategic investments to optimize our global footprint, and returning cash to shareholders within our capital allocation framework. We are confident that this multifaceted and balanced approach will allow us to create long-term shareholder value. Moving on to our Q2 outlook, we expect revenue between $1.375 billion and $1.475 billion, representing growth of 8% sequentially at the midpoint. We are closely monitoring the impact of tariffs and other trade regulations, which may affect demand. Gross margin is expected to be between 11.5% and 13.5%, reflecting a modest improvement in utilization across our factories. We expect operating expenses of around $125 million for the quarter and a full-year effective tax rate of around 20%. As a result, Q2 net income is expected to be between $17 million and $57 million, resulting in EPS between $0.07 and $0.23. Our CapEx forecast for 2025 remains unchanged at $850 million, of which 5% to 10% is estimated for our new advanced packaging facility in Arizona. Our investments are focused on expanding capacity and capability for leading-edge technology, including the next-generation RDL and bridge technology, advanced SIP, and test solutions. We will continue to monitor the environment and prudently evaluate our investments. Although tariff and trade regulations are constantly evolving, Amkor's operations remain largely unaffected. Our shipments to the U.S. are immaterial, and nearly all of our operations are in some form of free trade zone, which provides duty relief for the direct supply of equipment, materials, and components. We believe our diversified geographic footprint remains a competitive advantage, and we will continue to closely monitor developments in this area. In this dynamic environment, we remain poised and ready to support our customers. Our team continues to execute our strategic plan. We are investing in differentiated technology to maintain our position as a leader in the OSAT space. We are optimizing and expanding our geographic footprint, and we are closely partnering with leading semiconductor companies that are addressing market megatrends. With that, we will now open the call up for your questions. Operator. Thank you. At this time, we'll conduct your Q&A session. If you would like to ask a question, please press star one on your telephone keypad. A confirmation tone will indicate that your line is in the question queue. You may press star two if you would like to remove your question from the queue. For participants using speaker equipment, it may be necessary to pick up your handset before pressing the Star keys. Once again, to ask a question, press star one. We'll pause for a moment while we pull for questions. Our first question comes from Charles Shi with Needham. Please state your question. Hi. Thanks for taking my question. Maybe I missed a little bit the first part of the prepared remarks, but I wonder if you guys can provide some color on why the Q1 you've done a little bit better than expected, but Q2, I'm especially interested in your thoughts on Q2 because the amount of sequential growth for Q2, if I look at the midpoint of the guidance range, it's one of the stronger Q2 in your recent history. I wonder, do you see any potential impact from any pull-ins due to concerns on the tariffs for any of the two quarters, Q1 or Q2? Thank you. Hello, Charles. This is Giel. Let me start answering your question. First of all, in Q1, we had strength in our communication business, and the other businesses were performing in line with expectations. Now, for Q2, we see growth versus Q1, again in communication, and on top of that, we see strength in the computing domain, and that goes across applications from data center into the PC domain, but also in networking. That is a strength in Q2. Do we believe that the Q2 strength is related to market pull-ins? That is not what we observe, Charles. I think we see we are working with customers to optimize the utilization of our lines, but that is more on our own, let's say, management structure that we normally have, but it is not really a pull-in from our customers. We do not see that in neither of the markets that we serve. Got it. Thanks. The second question, you guys are maintaining that $850 million CapEx. Given the recent tariff news, any thoughts on the continued expansion in Vietnam and any thoughts that you may or may not continue the current investment in that specific country that probably can be hit by the tariffs? Yeah. Let me give you some color on our CapEx plans, and Megan can give more details later on, Charles. Currently, we do not foresee that we significantly change our CapEx plan, although we remain flexible as much as possible if we observe unexpected events later in the year where we then can push out certain investments. Now, 70% of our investments are in capacity and capabilities, of which, and then 25% is facilities and construction expansion, of which 5%-10% is for our Arizona and Portugal facility. Now, if we take the 70% on capacity and capabilities, it is mostly to support our high-performance computing market. That goes across different applications from PC into networking across 2.5D technology as well as RDL technologies. We still see a strong pipeline there, but that may be impacted by a tariff structure that may change in the course of this year. By watching that closely and working with our customers, we foresee that we have opportunities to, let's say, delay certain shipments of equipment, although we believe that long-term and midterm, this will continue to be a strong market, and most of our investments are for fungible equipment. On top of that, we invest in expansion of our test capability and capacity. Megan, anything to add from your side? Charles, I think as we look at our current view, our H1, I would say, shaped up slightly better than what we expected with our Q1 coming in at the high end of guide and then a strong compared to our seasonal expectations for Q2. As Giel mentioned, we're not seeing any significant changes from our customers' perspectives, so we are holding our CapEx expectations, although we will continue to monitor that. Thank you. Our next question comes from Ben Reitzes with Melius Research. Please state your question. Yeah. Hi. Good afternoon. Hope you can hear me okay. I wanted to ask about the communication segment beyond the 2Q. Should this be better than seasonal for the H2 given the win of the new socket? I was wondering what that would mean for margins. Also, in the answer to your question, if you've thought about tariffs and how that may impact the seasonality in the H2 as well in the comms sector. Thanks. It's difficult to give details on the H2 here, Ben, because what we can say on the H2 is that the fundamentals going into the H2 are unchanged. I mean, if we take critical programs, specifically in the communication segments, that's still in line with what we expected before. The new program with our lead customers, we have line of sight, and we will start initial ramp in the Q2 towards the end of the Q2. We expect that to happen. Our projections and our program and project pipeline in the communication segment is also expected. We believe that our market position there is strong, both on the iOS side as well as on the Android side. However, given the trade restrictions and the uncertainties there, the macro indicators like the consumer demand profile and impact of, let's say, tariffs may impact the total volumes and may influence bill plans of customers. That creates significant uncertainties. Now, for ourselves, we clearly focus on the elements that we can influence, and we're very confident that our position in the communication market is as we predicted earlier. Ben, just to comment on your question with respect to gross margins. As we move into the H2, we would expect those gross margins to expand as we increase utilization. That would be expected given our financial model. As Giel mentioned, we'll be watching demand carefully in order to flex costs where possible to manage profitability. Okay. Great. If I could just slip in one more, I wanted to just clarify on the auto industrial guidance, given the margin impacts that you have in that vertical. What are you thinking there? Can you just reiterate what you said about the two Q and in the H2? And does that have the same uncertainty with regard to tariffs as the other spaces, or is that space really picking up? We believe that the automotive markets and the performance in that market hit the trough, and we are very careful in anticipating further growth. Now, we definitely see strength on the advanced packaging in automotive that is driven by in-car infotainment, ADAS applications, and that comes also with peripheral devices like image sensors and other sensor devices. We see strength there. On the more mainstream side of the automotive market, if we talk to our customers, the lead customers there are confident that they hit the trough, although we do not see too much of a growth signal there going into the next quarter. Overall, we believe that the Q2 is giving, let's say, a single-digit, mid-single-digit growth quarter on quarter, but still, we are careful in predicting a strong H2 of this year in automotive. Thanks a lot. Appreciate the answers from both of you. Your next question comes from Randy Abrams with UBS. Please state your question. Yes. Hi. Thank you. I wanted to ask that first question on the Arizona expansion. Two parts to it. F irst, how you're thinking about your opportunity now with TSMC announcing their two sites in the U.S. If you could talk where you see your opportunity versus their expansion. The other side, it sounds like even given that you're expanding or potentially considering a faster plan. If you could talk the potential to either pull in or raise the scale of Arizona if indeed you see a bigger opportunity there. Yeah. Good point. Let me share our view there. Now, the announcements made by TSMC for enhancing and increasing their investments in the U.S., we believe that's an opportunity for Amkor. The increased scale of their operation, also the enhancement of next-generation technology that would come into the U.S. will mean that the volumes will increase. We currently believe that in our strategy to be complementary to TSMC, that that will increase the opportunity for Amkor. Now, we're currently evaluating the technology portfolio for our Arizona facility because that will be impacted, moving maybe more to an on-substrate technology, but that's still to be seen. We are looking at two ways currently. One is to accelerate, and secondly, to scale up faster than we originally anticipated in line with increasing demand. That's how we currently evaluate the opportunities in the U.S. Randy. Okay. Yeah. A quick follow-up, and then I'll ask a second question. Faster than anticipated, if you could give the timeline how you see the scale-up in terms of going into volume for that facility. The second question on a couple of the advanced technology, curious how you're seeing the potential come in. I think you mentioned the RDL interposer is accelerating some of the R&D. How you see in terms of design activity or opportunities once you go into production for that. The second, you mentioned a bit more on the co-packaged optics, when you see timing and how you see participating versus TSMC, which has some vertical integration there. Okay. Let me start with the first part of your question related to the timeline, Randy. We are currently still on schedule to start our manufacturing location in Arizona, the building in the H2 of this year. We are looking for ways to accelerate this, and I think options are there to have a 7 x 24 building schedule. You have to keep in mind that building a packaging facility for us, our last facility that we built in Vietnam, took us 18 months from groundbreaking to equipment move-in. We expect that to be a little bit longer than Arizona, but we are looking for ways and means with our contractors to accelerate that. Now, also with respect to the scale and the technology portfolio, now, originally when we started to evaluate an opportunity, and that was also publicly announced with TSMC and our lead customer, we're very much tuned towards starting with communication technology. Currently, with the compute market growing as it does, we're evaluating the best technology portfolio there. We believe that it's a combination of communication together with compute technologies. Compute would be a combination of a few elements. It would be an RDL or 2.5D technology portfolio combined with an on-substrate portfolio. We're working with TSMC to be as seamless as possible for customers that transfer their product portfolio out of Asia into the U.S.. We currently anticipate that that will be a portfolio of technology that we will support. Now, on the timing exactly of I cannot talk about on behalf of TSMC, so there's still many details to be discussed, and we are in constant communication to get the details sorted out. Now, on the co-packaged optics, we currently work with a lead customer, which is a leading data center company on the first products that go into the market. We're actually in production now. There are multiple new generations coming up, and that will diversify. We see, let's say, the co-packaged optics technology to diversify both within the data centers as well as within the networking. We expect multiple devices to ramp over the next couple of years. Thank you. Your next question comes from Craig Ellis with B. Riley Securities. Please state your question. Yeah. Thanks for taking the question. In fact, I'll start with a couple of clarifications. If you could clarify in the communication segment, you mentioned it was stronger than you expected in the Q1. Can you just characterize the linearity of that strength? Was it all quarter-long, early quarter, late quarter? In the RDL-based opportunity you just talked about, can you clarify, is that something that as you work on it this year generates revenue this year, or is that revenue-generative next year? Can you comment if that would be first or H2? Very good, Craig. With respect to your first part of the question, how the communication segment developed and how the improved revenue developed over the quarter, I don't have the data here, but I would expect that it would be rather linear. We started the quarter pretty strong, and it developed in line with normal expectations throughout the quarter, but a bit stronger than expected. We have a weekly review of how our revenue expected, but now from the beginning of the quarter, we saw, I would say, a fairly solid communication business build-up. With respect to the RDL technology, currently we have one device in production for a CPU data center device. There are multiple other devices currently in qualification, and we definitely think that that's a next-generation technology that will take off. The investments will result in revenue in this year. You have to keep in mind that the investments that we're making for high-performance computing are fungible across applications. The capacity that we put in place for 2.5D, for RDL, for large body-sized Flip Chip BGA, even for co-package optics, the equipment instances are fungible across the whole domain, even for standard bumping. We are able to reach a high utilization. Most of the investments will start generating when they come in. Of course, we need to qualify the equipment, but we'll start generating revenue towards the end of the year or early next year. That's very helpful. Thank you. The follow-up is really a higher-level question for you or Megan. It goes back to some commentary from last quarter's conference call. I think in characterizing the year, the company was of the view that revenues could be flat to modestly up year-on-year with strong half-on-half seasonality, potentially in the 40-60% range, half-on-half. Is that still the right way to think about the year, or how should we think about the year's potential and the way linearity could break down? I hand that over to Megan. Hi, Craig. Yeah. As I had just mentioned, our H1 has shaped up better than we had originally expected. That alone would moderate some of that H1, H2 magnitude that we had talked about last quarter. With respect to our position on the H2, we're not providing H2 or full-year guide given the dynamics and the environment. That being said, the fundamentals of what's driving our second-half growth are still intact. Those are really centered around communications with that new socket and the seasonal launch expected, a lot of the new compute programs that Giel has mentioned, strength in automotive, as well as the continuation in consumer for that wearable product. With that, we're just going to leave it at that point given the dynamics. Thank you. Our next question comes from Peter Peng with JPMorgan Chase. Please state your question. Hi, guys. Thanks for taking my question. I want to double-click on the computing and specifically your 2.5D segment. It sounds like you guys are just given some of the expert controls at your marquee customers. You're downticking that, but you're also ramping with a networking switch customer this quarter. Given the puts and takes, should we expect that this segment could be flagged for the year, or just given that marquee customer is quite big, that this could be declining this year? Yeah, Peter, let me start by sketching the color in the compute segment. Now, as you mentioned, we are indeed ramping up with the second customer in 2.5D, and we are continuing with 2.5D for our prime customer, although at lower volumes due to the current situation on export controls. Next to that, we are ramping up multiple other devices for both customers, both on the GPU domain as well as the CPU domain, and we expect that to be in production for the H2 of this year, given the uncertainties of the overall environment. Next to that, we have multiple devices in qualification that will have strength in the H2 of the year. That goes beyond the wafer-based technologies. There are certain devices in the data center that go for large body-sized Flip Chip BGA. All of that contributes to strength in the computing domain. We're optimistic for the H2 of this year, for the full year, given the disclaimer that in the current macro environment, things may change rapidly. Got it. Maybe just kind of follow-up to that question, maybe as we kind of look into 2026, I know you guys were working on the Connect- S Technology with the CoWoS-L. Any updates on there and whether you're able to support that marquee customer for next year? Maybe more broadly, if you can talk about your design wind pipeline for your 2.5D segment. Let's first try to look at the first part of the question on the bridge technologies. Now, currently, we have two customers in qualification to be ramping by 2026. In the course of next year, the volumes are hard to predict, but we expect that to complement our 2.5D and RDL-based portfolio. We still expect growth, and that's also how we invest. We also see diversification in the compute segment where multiple devices, as well as in the data centers, are moving from GPU. On top of that, we see CPU programs in data centers. We also see networking devices being refreshed and renewed in data centers. The portfolio is broadening beyond, I would say, the single device, GPU device that you're referring to, Peter. Thank you. Our next question comes from Steve Barger with KeyBanc Capital Markets. Please state your question. Thanks. As you move into the socket recovery ramp, does the content and pricing there mean you'll outgrow the prior program? Do you have visibility into share gains or content gains beyond the socket issue that we should be thinking about, or is whatever happens in the back half just more about unit volume? Let's start from the latter part of your question. Our overall business in communication is very dependent on unit volume. Unit volume depends on build plans for individual customers, both in iOS as well as in the Android ecosystem. These build plans may be impacted by current trade restrictions, export controls, etc. That as a disclaimer. Now, to go back to your earlier part of your question, how is the socket that we're currently ramping in the latter part in the next generation iOS devices? How does that compare to previous devices when it comes to technology, complexity, and pricing? If we compare it on revenue potential for the H2 of the year and we compare it with previous devices, then it's in the same order of magnitude. The devices are different, but the total revenue potential is in the same order of magnitude if the volumes are the same. That is the uncertainty that we have. In the middle part of your question, what is our view? What is our visibility of our, let's say, position in the individual phones? Do we have a clear visibility of our socket position beyond the sockets that we regain? There we have a very open communication with our lead customer and customers, both in iOS as well as in the Android ecosystem, and we know pretty exactly which positions that we're in. Slight uncertainty is that in some of the positions and slots, there are multiple suppliers. It is a bit on market share division. Overall, we know exactly what our footprint is, and we do a very detailed planning on volume expectation for the ramp of a new phone generation. Understood. That's good detail. Thank you. I know you can't make firm statements about trade issues at this point, but if you do have customers asked to move a program to mitigate tariffs and it goes into an area with higher labor costs, or if you incur costs because you have to move things around, is that all complete pass-through, or how will that customer conversation go? This goes on a case-by-case basis. We cannot say that we have a general approach. I mean, we are all in a, I hope, once-in-a-lifetime experience, but this is definitely a first for many of us. We do that on a case-by-case basis. I mean, generally, we work with customers for many, many years. I think on our lead customers, generally all more than 10 years, we have a conversation, and there is a certain rationale that if we incur more cost, that we look at the rationale and we come to an agreement. I feel that in the current way that Amkor approaches this, we work with a high level of agility, and we see our customers work in the same way, looking for practical and pragmatic solutions in case that bottlenecks occur in the supply chain. Understood. Thank you. Thanks. Our next question comes from Thomas Diffely with D.A. Davidson. Please state your question. Yes. Good afternoon. Thank you for taking my questions. Giel, curious to maybe follow up on the communication side. Do you expect the AI going to the edge to drive unit growth this year in the handsets? And perhaps if not, is there enough technology changes to allow you to grow in a further depressed market? That's not an easy question to answer, Tom. With respect to AI entering the smartphone segment, we believe it will enter the smartphone segment through the premium tier of the market. That's basically good for Amkor because that's where we are concentrated when it comes to our market position. Will it drive upside this year already? It's difficult for me to forecast that part of the market. Will it come at a given point in time? I currently see that there is definitely a belief in the market that AI will proliferate towards edge devices, whether it's smartphones or whether it's PCs. I think there are some first applications coming on the market. It will drive innovation in the smartphones, whether it's innovation in the connectivity part, whether it's innovation in the compute or co-processor part in the phones or in the modems. I think many of these devices will go through an innovation cycle. When will it come? When will it start driving upside? It's difficult to say. This year is a very difficult year anyhow for the market to predict volumes. Okay. No, that's fair. Megan, I had a question on the margins for the reported quarter. When you compare fiscal Q1 2025 to fiscal Q1 2024, on lower revenue, you had about the same COGS. I'm curious, obviously, you talked about overhead absorption being an issue, but to get the same level of COGS, was there a difference in the mix, or was there weaker pricing, or how would you describe what drove that? Tom, I just want to clarify. You're looking at Q1 2025 compared to Q1 2024? Correct. Yes. Really, the main difference between those two periods is moving our Vietnam factory to production. Those costs are having an impact on the margin in 2025 that was not there in 2024. For Q1, that was around 100 basis points. Okay. No, that's perfect. All right. Thank you. Thank you. At this time, I'm showing no further questions. I would like to turn the call back over to Giel for closing remarks. Thank you. Now, let me recap the key messages. First quarter results met our expectations with revenue of $1.32 billion and EPS of $0.09. Second quarter revenue is expected to grow 8% at the midpoint of guidance of $1.425 billion. We stay agile to support our customers in this dynamic environment, and we stay focused on executing our long-term strategy by strengthening our technology leadership, expanding our geographic footprint, and partnering with lead customers in growth markets. Thank you for joining the call today. Ladies and gentlemen, this concludes today's conference call. You may disconnect.