Himax Technologies, Inc.

Hello, ladies and gentlemen. Welcome to Himex Technologies, Inc., second quarter, 2026 earnings conference call. At this time, all participants are in the listening mode. And later, we will conduct a question and answer session. And instructions will follow at that time. And as a reminder, this conference call is being recorded. I would now like to turn the conference over to Ms. Karen Tiao, head of IR, PR at Himex. Ms. Tiao, please go ahead.

Welcome, everyone.

My name is Karen Tiao, head of IRPI at Hanex. Joining me today are Jordan Wu, President and Chief Executive Officer, and Jessica Pan, Chief Financial Officer. As our companies prepare comments, We have allocated time for our questions in the Q&A section. If you have not yet received a copy of today's result release, please email hf-ir at hymex.com.tw or himx at mcgroup.us or download a copy from Hymex's website. Before we begin the formal remarks, I would like to remind everyone that some of the statements in this conference call, including statements regarding expected future financial results and industry growth, are forward-looking statements that involve a number of risks and uncertainties that could cause the actual event of results to differ materially from those described in the conference call. A list of risk factors can be found in the company's latest SEC filing. foreign 20F in the section titled Rates, Factors, as may be amended. Except for the company's full year of 2025 financials, which were provided in the company's 20F and filed with the SEC on March 27, 2026, the financial information included in this conference call is unaudited and consolidated and prepared in accordance with IFRS accounting. Such financial information is generated internally and has not been subjected to the same review and scrutiny and may vary materially from the audited consolidated financial information for the same period. On today's call, I will first review Himex consolidated financial performance for the second quarter 2026, followed by our third quarter outlook. Jordan will then give an update on the status of our business and after which we will take questions. You can submit your questions online through the webcast or by phone. We will review our financials on an IFRS basis. The rapid rise in AI demand is placing unprecedented strength on memory chip supply and affecting many non-AI applications, creating a more challenging cost and capacity environment across the industry. Against this backdrop, we are pleased to report that our second quarter revenues, gross margin, and profits all exceeded the guidance we provided on May 7, 2026. Second quarter revenues registered $227.4 million, representing a sequential increase of 14.2% and up 5.9% compared to the same period last year. Q2 revenues exceeded our guidance range of a 10 to 13% increase, primarily driven by better-than-expected automotive IC sales.

Cost margin was

33.1%, substantially exceeding the guidance of around 32%, up from 30.4% in the previous quarter and 31.2% a year ago. This is primarily due to a more favorable product mix, which increased the sales for higher-margin automotive IC products. Queue-to-profit per diluted ADS was 11.4%. significantly exceeding the guidance range of 8.6 to 10.3 cents, up from 4.6 cents in the previous quarter and 9.5 cents a year ago. Revenue from large display drivers came in at $19.2 million, representing a decline of 21.0% from the previous quarter, attributable to panel makers pulling forward their inventory purchases for high-end TV ICs in prior quarters. In contrast, sales for both the monitor and notebook IC products increased quarter over quarter due to higher legacy product shipment to key customers. Sales of large-panel driver IC accounted for 8.4% of total revenues for the quarter, compared to 12.2% last quarter and 11.6% a year ago. Revenue from small and medium-sized display driver IC segment totalled $162.3 million, reflecting an increase of 19.6% sequentially. Q2 automotive driver sales, including both traditional DDIC and TDDI, increased by double-digit quarter-over-quarter, primarily driven by broad-backed customer replenishment of TDDI and DDIC following seasonally lower shipment during the new year in Q1. The run-up of new TDDI and DDIC projects for leading panel customers also contributed to a sequential increase. customers continue to operate under a make-to-order model while maintaining lean inventory levels. Our automotive business, comprising DDIC, TDDI, T-Con, and Olay IC sales, remain the largest revenue contributor in the second quarter, representing well over 50% of total revenues. Second quarter tablet IC sales, covering both LCD and Olay products, also increases sequentially, attributable to customers' early pull-in demand against the backdrop of the rising memory price sentiment in the market, together with the continued shipment for a customer's premium OLED model. In contrast, smartphone IC sales decreased sequentially following the initial roundup of an OLED IC for a leading smartphone brand's mainstream model in Q1. The small and medium-sized driver IC segment accounted for 71.4% of total sales for the quarter, compared to 68.2% in the previous quarter and 67.3% a year ago. Q2 non-driver sales rate. A $45.9 million, a 17.7% increase from the previous quarter, attributable to robust automotive tea consumption, supported by a reflection across the board customer base. Teacup business accounted for over 10% of the total sales, with more than half contributed by automotive teacups. As the market leader in automotive techon, particularly in solutions featuring local demifunctionality, we expect strong growth momentum to continue into next year. Non-driver products accounted for 20.2% of total revenues as compared to 19.6% in the previous quarter and 21.1% a year ago. Second quarter operating expenses were $15.7 million. and increases of 0.8% from previous quarter and 3.6% compared to the same period last year. The year-over-year increase was mainly attributable to higher tap-out expenses. We remain disciplined in managing cost while continuing to invest strategically in Fulagnum's rather icy business with compelling long-term growth potential. Second quarter operating income was $24.6 million, representing an operating margin of 10.8%, compared to 5.1% in the previous quarter and 8.4% for the same period last year. Both the quarter-over-quarter and year-over-year changes were primarily driven by higher revenues and gross margins. The second quarter after tax profits was $19.9 million, or 11.4 cents, per diluted ADS, compared to $8.0 million, or 4.6 cents, per diluted ADS last quarter, and up from $16.5 million, or 9.5 cents, in the same period last year. Turning to the balance sheet, we had the 298.8 and other financial assets as of June 13, 2026. This compared to $332.8 million at the same time last year and $287.6 million a quarter ago. The sequential increase was mainly driven by operating cash flow of $17.5 million in the second quarter. Before moving on, I would like to highlight one point regarding this quarter's cash flow. This is our usual practice. Income tax payments are made in the second quarter. Under a new Taiwan government policy, we are entitled to defer approximately $11.0 million of this payment for one year without interest. Including these deferrals, second quarter operating cash flow would have been approximately $6.5 million. Looking ahead to Q3, we anticipate a decline in cash, cash equivalent, and other financial assets, primarily due to the payment of the $44 million for the annual dividend to shareholders made on July 10th. In addition, subject to the final vote decision, we will distribute around $11.7 million, the immediately best portion of this year's employee bonus award at the end of the Q3. Our quarter end inventory as of the June 13th, 2026, were $151.5 million. filed the same as the $151.7 million on last quarter, but higher than the $134.6 million in the same period last year. After maintaining lean inventory levels for several years, we proactively adjusted our inventory strategy about a year ago, selectively building inventory in anticipation of the tightening supply across the industry. Account receivable at the end of the June was $220.3 million, up from $190.9 million last quarter and $219.0 million a year ago. DSO was 93 days at the quarter end, as compared to 86 days last quarter and 92 days a year ago. Second quarter capital expenditure primarily for R&D-related equipment for our IC design business was $4.3 million versus $2.9 million last quarter and $4.6 million a year ago. As of June 13, 2026, Himex had a 174.4 million ADS outstanding on change from last quarter. On a fully diluted basis, the total number of ADAs outstanding for the second quarter was 174.4 million. During the quarter, on July 1st, we announced the proposed divestiture of investment in one of our equity method investor keys. Based on the information provided by the said investor key company, We expect to recognize a pre-tax gain of approximately $23 million to $24 million upon closing. The transaction is expected to close in the fourth quarter of this year, subject to customary closing conditions and regulatory approval. We will provide more updates as appropriate as the transaction progresses. Now turning to our third quarter 2026 guidance. We expect Q3 revenue to increase the 7% to 11% sequentially. Cost margin is expected to be around 34% depending on the product mix. Q3 profit attributable to shareholder is estimated to be in the range of the 8.0 to 10.0 cents per fully diluted ADA. As we have done historically, we will grant employees annual bonus, including RSUs and cash awards, on or around September 13th this year. The third quarter guidance for profit per diluted area has taken into account the expected 2026 annual bonus, which, subject to board approval, is now estimated to be around $13 million. out of which $11.7 million will be vested and expensed immediately on the grand stage. As a reminder, the total annual bonuses amount and the immediately vested portion are our current best estimate only, and the actual amount could vary materially depending on, among other things, our Q4 profit expectation and the final board decision for the total bonus amount and its vested scheme. It is also worth noting that the $30 million expected annual bonus does not yet include the above-mentioned gain on investment from diversification of the equity method Invest-T. The transaction is pending regulatory approval and has not yet closed. Thank you very much. Thank you. In comparison, the annual bonus for 2025 and 2024 were $7.7 million and $12.5 million, respectively, of which $7.5 million and $11.2 million vested immediately. In providing our Q3 financial guidance, the Q3 expense related to the employee bonus is estimated to be $11.8 million, representing 6.8 cents per diluted ADS before tax, comprising of the $11.7 million of the immediately vested portion of this year's bonus as stated above. and $0.1 million of the amortized portion of the unvested bonuses from previous years. By comparison, employee bonus expenses in each of the last three quarters was around $0.2 million. I will now turn the call over to Jordan to discuss our Q3 outlook.

Jordan, the floor is yours. Thank you, Karen.

The ongoing surge in air demand continues to impact non-air applications. This has rippled across the broader semiconductor supply chain, resulting in capacity constraints as boundary packaging and testing facilities on the material process nodes where many of our products are manufactured. Consequently, we are experiencing We expect the supply environment to remain challenging in the near term. To enhance production accessibility, and secure the capacity needed to meet our customer needs and support our upcoming production ramps. We continue to leverage our established supply chain in Taiwan, while further strengthening our presence across China, Singapore, Korea, Japan and Malaysia. In parallel, As we mentioned last quarter, we have been working closely with customers on pricing adjustments to share these increased costs. Some adjustments took effect in the second quarter, with additional pricing adjustments possibly implemented on time as the market conditions warrant. Now, withstanding these industry-wide supply constraints, We remain optimistic about the long-term growth prospects of our automotive display IC business. We continue to view automotive as one of the industry's most attractive secular growth markets, driven by rapid advancements in smart vehicle materials. This trend is characterized by a number of other things, A growing number of displays per vehicle, now averaging more than three and continuing to rise, along with larger, higher resolution displays and more diverse vehicle cabin configurations, including curved, integrated, multi-display and pillar-to-pillar designs. Timex is well-positioned to capitalize of Automotive Display Portfolio, spanning both LCD and OLED technologies. A broad and diversified global customer base and a robust design we pipeline. We further differentiate ourselves by continuously introducing next-generation automotive display technologies, including LTDI solutions for ultra-large displays Advanced T-Con Solutions for Head-Up Displays, OLED Driver and Touch Control ICs, and Micro-Reality Display Technologies. Our portfolio also includes platform display solutions and capacitive physical buttons, where customer interest continues to grow, driven in part by regulatory and vehicle safety initiatives in key automotive markets, including China and Europe, where greater emphasis is being placed on intuitive physical controls to enhance driving safety and reduce driver's distraction. The industry's ongoing pursuit of richer human-machine interfaces, immersive infotainment and enhanced in-cabin user experiences is driving the adoption of a broader range of our display technologies. This not only increases HIMAX's dollar content per vehicle, but also creates multiple long-term growth opportunities. In addition, our well-established global supply chain provides us with greater flexibility to navigate the current supply environment We also see an important trend in the automotive industry, with automakers introducing new vehicle models at an accelerated pace and intensifying competition. As a result, product life cycles are becoming shorter, creating greater pressure to improve engineering efficiency, reduce development costs, and shorten time to market for new car models. These challenges are driving broader adoption of platform standardization across multiple vehicle models, favoring suppliers with comprehensive and validated technology portfolios and Proven Track Records. Himex is one such supplier, hosting the industry's most compelling automotive display IT offerings, market-leading positions across automotive BBIC, TDDI and TECON, and continued leadership in next-generation technologies such as LED and OLED technologies. by adopting hybrid solutions as part of their standardized platforms. Customers can quickly deploy validated display IT products across multiple new vehicle developments, reducing engineering efforts, lowering system costs, and shortening development cycle of each project. In addition to automotive, We are also making solid progress across several strategic growth areas, including smart glasses, artificial power and AI, and CPO. These emerging businesses diversify our revenue base into markets with attractive long-term growth prospects and margin profiles while strengthening our overall competitive position. We believe they are poised to become increasingly meaningful contributors to our future growth. First, on smart glasses, an area we remain particularly optimistic about. Hymax is one of the few companies offering both ultra-low-power AI sensing and micro-display technologies, both critical building blocks for next-generation smart glasses. On the AI sensing front, WiSight enables ultra-low power always on outward and inward sensing, supporting an expanding range of AI use cases, including contextual awareness, real-time visual assistance, and personalized user experiences. Recently, a leading global brand just launched a smart glasses product powered by our Wi-Fi technology, and we continue to see strong design momentum across customers worldwide. In particular, we are seeing growing engagement with leading global brands, technology platform providers, ODMs, as well as hyperscalers, who traditionally do not offer hardware products, but are now entering smart glasses market, with some projects poised to enter mass production as we move into 2047. On the display side, our Fronit Aircos micro-display delivers an optimal balance of size, weight, resolution, image quality, power consumption, and cost. It can also be configured to operate in the high-brightness ultra-low-power green-only mode and seamlessly switch to full-color operation as needed. This flexibility differentiates our solution from alternative display technologies, letting customers optimize power efficiency while maintaining visual performance and meeting their system design cost targets. Together, these capabilities make our front-end vehicles a compelling display platform for natural-version AI glasses equipped with see-through displays. Currently, we are working closely with multiple web-guided partners across Asia, Europe, and North America to deliver integrated AI display solutions that simplify system integration and shorten customers' This is driving broader customer engagement and positioning us to convert more opportunities into design wins. Backed by well over a decade of Airco's expertise and a proven track record of successful production shipments, we are well positioned to support the next generation of AR glasses. Now I would like to provide a brief update on our progress in CTO. Customer development and timelines remain aggressive, with demand showing no sign of slowing. Together with our strategic partner, 4C, we continue to deepen customer engagement by offering a flexible portfolio of solutions, including customized designs tailored specifically for our customers' needs, as well as a standardized technology platform developed in collaboration with a leading foundry partner. Our solutions support both co-packaged and pluggable packaging architectures to address diverse customer needs. Our primary focus for the second half of this year remains on achieving mass production readiness, including key customer qualification milestones, while continuing to improve manufacturing yields and establish stable mass production capabilities. We have already made encouraging progress towards these objectives. Both our Gen 1 products and Gentle Product, designed for 6.4T bandwidth, have begun engineering production ramps as scheduled in the third quarter. These products are expected to drive sequential shipment growth quarter over quarter, laying the foundation for a more meaningful volume shipment beginning in 2037. The official mass production timing remains subject to customer deployment schedules. Nevertheless, we expect our shipments in 2027 to be significantly higher than in 2026, starting to make meaningful contributions to our financials. At the same time, we are also co-developing next-generation optical solutions with customers, featuring higher Hypercount Architectures, Enhanced Optical Precision, and increasingly sophisticated designs such as CWDM or Coarse Web Lens Division Multiplexing to address the explosive bandwidth demands of WPC and AI data center applications. With that, I will now begin with an update. on the large panel drive IT business. In Q3, large display drive IT sales are expected to decline by single digit from last quarter. Monetary IT sales are expected to decline quarter over quarter as customers already pull forward inventory purchases in prior quarters. In contrast, TVIC sales are poised for sequential increases, driven by higher legacy product shipments to key customers. Looking ahead in the notebook market, we are seeing encouraging design momentum, particularly in OLED notebooks, with several industry trends creating favorable tailwinds. Writing memory prices are accelerating the shift from entry-level to premium notebook models. Prior to the scheduled wrap-up of new Gen 8.6 OLED effects in China later this year and in 2037, it is expected to further drive OLED penetration in the notebook market. Timex is well-positioned to capitalize on these favorable industry trends with our comprehensive notebook supply, notebook display, all the IC portfolio, spanning DDIC, TCAM, power management IC, and touch controllers. This one-stop offering lets us and others. In Q3, small and medium-sized display driver IC business is expected to increase by a higher single digits from last quarter. Q3 automotive driver IC sales including TDDI and traditional DDIC are set to increase by a solid double-digit quarter-over quarter. This increase reflects broader customer demand of DDIC and TDDI products, together with the mass production of multiple zero-TDI projects across car brands worldwide. Strong sequential growth underscores The accelerated adoption of larger and more sophisticated automotive displays, which Chinese automakers needed the charge. We continue to see healthy underlying customer demand, supported by multiple new projects and three mass production in the coming quarters. We expect our full-year 2026 automotive driver IC sales to grow by double digits from last year, with strong growth momentum extending into next year as adoption of smart car interiors continue to drive increases in the number, size, and sophistication of displays in both electric

and conventional vehicles.

As I noted earlier, the industry's shift towards platform standardization is creating meaningful opportunities for Himex. This is evidenced by the growing number of customers who adopted our industry-pioneering IoT DEI-level DMT-Con solutions as the standard platform for their ultra-large Automotive Displays. Following years of customer engagement, several of these projects are now entering mass production across multiple car brands. These ultra-large display panels typically require four or more LTDI chips, and in some cases more than ten, together with at least one double-dimming T-Comp per panel. As customers increasingly adopt our solution across multiple ultra-large display platforms, this not only strengthens customers' stickiness and makes it more difficult for competitors to compete with us, but also increases our content value on a per-panel and per-vehicle basis. Looking ahead, The accelerated adoption of OLED displays in automotive presents a compelling long-term growth opportunity and is poised to become a key pillar of our automotive business. For several years, we have been collaborating closely with leading OLED panel makers in Korea and China, and our comprehensive portfolio of TDICs, TCAPs, Punch Control ICs, and customized ASIC solutions, gives customers the flexibility to select the solutions that best meet their specific requirements. This broad product coverage and our early customer engagements have already translated into numerous development programs providing a solid foundation for future growth as premium automotive displays transition from LCD to OLED. With OLED adoption on the way, we continue to introduce innovative IT solutions to address evolving customer needs. For example, our latest TED or TECON embedded driver IC solution, which integrates A cost-effective, flexible, and highly integrated solution ideal for smaller, lower resolution automotive displays. Our TED technology is now being adopted across a diverse range of applications, including automotive, robotics, and IT applications, with several projects involving customized basic solutions co-developed with leading global and customers. Moving to smartphone IT sales, we expect Q3 smartphone revenue to increase quarter-over-quarter through a live continuous shipment for leading smartphone brands, mainstream models, and inventory build-up for its upcoming premium models. For tablet ICs, Q3 sales are expected to decrease sequentially as capacity constraints limit our ability to support additional shipments. I would like to now turn to our non-drive-by-sea business update, where we expect Q3 revenue to increase by low teams sequentially. First, for an update on our TCAM business, We anticipate Q3 T-Camp sales to increase by 12% quarter-over-quarter. Our automotive T-Camp business is expected to deliver decent double-digit growth in Q3, extending the strong momentum from Q2 and far outpacing our corporate average. This growth is driven by continued legacy product shipments across a broad, diversified customer base. along with several new projects entering mass production. Despite ongoing industry-wide capacity constraints, we are committed in the strong growth trajectory of our automotive teacup business. With hundreds of design wins already secured and new design means continuing to expand, we are well positioned for another robust growth year in automotive teacups as we move into . During the quarter, we are pleased to announce that our T2000 TCAM has been adopted into E-Ink's next-generation coloring paper platform. PIMAX proprietary parallel processing architecture is at the core of this ASIC product, enabling simultaneous display refresh and Data Transmission, significantly enhancing dynamic display performance while preserving the ultra-low power advantage of ePaper technology. This breakthrough enables smoother display of dynamic content on large-format ePaper displays, helping accelerate the transition from traditional stacked signage to dynamic applications such as retail advertising, public information displays, and smart commercial environments. Switching gears to the Wi-Fi product line, a cutting-edge ultra-low-power AI sensing total solution targeting battery-powered endpoint devices Wi-Fi differentiates itself with an industry-leading ultra-low-power architecture, consuming only a few milliwatts while delivering on-device AI inferencing, 24-7 always-on image and voice sensing, and an exceptionally compact form factor. This unique combination enables endpoint AI devices that were previously are all impractical due to power and size constraints. Drawing broad adoption across applications including notebooks, surveillance systems, SS control, PalmVan authentication, smart office, and smart glasses. With design activities continuing to expand across leading customers worldwide. On the YSI module's front, design activities continue to expand, driven by its plug-and-play architecture, ultra-low power consumption, and all-device AI capabilities. During the quarter, we were pleased to announce that our YSI biometric power band modules achieved the certification of TUV Redland. One of the world's leading and most credible independent testing, inspection, and certification organizations. These assessments validated our recognition accuracy, response speed, and likeness detection. This verification, together with our early achievement of GDPR compliance, one of the world's strictest data privacy standards, reinforces the privacy, security, and performance of our biometric authentication solutions, giving customers greater confidence to accelerate development across security-sensitive applications. We are seeing expanding design-in activities from our PowerMed modules across smart access workforce management, smart door locks, and more, and more recently, computer monitors and smart office solutions. Built on the same core hardware platform as the Wi-Fi technology, our Wi-Scar module is specifically designed for security applications delivering ultra-low power operations for wide field of view, long-range detection, and exceptional low-light performance. WhiteScar accurately detects and continuously tracks multiple individuals, including their presence, location, and movement, substantially reducing the false triggers commonly associated with traditional PR-based solutions. Its proactive 24-7 sensing capability enables security systems to detect and continuously track activity from the outset, capturing the full sequence of events rather than only the moment motion is detected. providing a significant advantage over traditional reactive solutions. White Scar delivers up to five years of battery life while maintaining high precision detection over long distances, even in environments with illumination as low as one lux. Since its debut just six months ago, WhiteScar has seen encouraging customer engagement across a wide range of applications, including surveillance cameras, access control, IoT, and wildlife monitoring. We are also pleased to share that WhiteScar has already been adopted by a US customer for surveillance applications, with mass production scheduled to begin towards the end of this year. As mentioned earlier, WISE-EYE is gaining broad market recognition for smart glasses as a compact ultra-low power always-on AI perception front-end. WISE-EYE supports our phasing environmental sensing, first through scene understanding by analyzing the user's contextual surroundings and environment. followed by object classification to recognize and identify specific objects physically associated with the identified scene. We also support mirror-facing capabilities, including eyeball tracking for intuitive gaze-based interaction and iris authentication for secure identity verification. Together, these capabilities enable AI glasses to continuously capture visual snapshots of the real world and enable intelligent, responsive, low-latency human-machine interaction while consuming only a few milliwatts of power. With the leading global brand launching wide-eye-powered smart glasses this fall, we are seeing growing engagement from platform providers, ODMs, and consumer electronics companies worldwide. Some of these projects are expected to enter mass production in the coming quarters. That concludes my report for this quarter. Thank you for your interest in HIMAX. We appreciate your joining today's call and are now ready to take questions.

Yes, thank you, Jordan. And ladies and gentlemen, we are now in question and answer session. If you would like to ask the question, please press star key and number one on your telephone keypad and you will enter the queue. And after you are announced, please ask your question. If you find that your question has been answered before it's your turn to speak, you may press star key and number two to cancel the question. And in addition to submitting questions via phone, You may also submit questions through the webcast, where the checkbox is available on the right-hand side of the screen. Thank you. Now for dialing participants. If you would like to ask questions, please press Start 1 on your telephone keypad. Thank you. And our first question will be coming from Donnie Tan Nomura. Go ahead, please.

Thank you, Jordan, for taking my question. My first question is regarding to the CPO revenue outlook into 2027. As you mentioned about the revenue could be more meaningful for next year. Can you classify how do you define meaningful for 2027? And maybe you can also provide us with the overall business outlook in 2027 as well. Thank you.

Thank you, Tommy. Perhaps let me start with your second question that I'll look for next year. As you know, we don't usually provide guidance beyond one quarter, but we are positioned, I think, to see robust sales growth for next year. which continuously improves gross margin. As you know, our gross margin this year has improved from last year, and we believe the trend is likely to continue next year. Before I elaborate further, there is one important thing for clarification, and as you mentioned in your question, Tony, we are going to, for the time being, before CPO starts mass production officially, It's a good idea that we kind of separate our CPO for our Outlook discussion because our CPO is arguably a once-in-a-lifetime opportunity for Hymex at the pre-mass production stage. Its Outlook will be too uncertain to quantify, at least for the purpose of Outlook comments. So what I'm about to say below does not get the SNAP-included contributions from CPO. And I will leave the comments on CPO for the next question, which you already just raised. So we start now back to our view on 2027. I think a very important trend for next year, our financials, is that the non-driver areas will likely outgrow driver ICs, with revenue contribution approaching 30% from around 20% at present. This is driven mainly by automotive key customers, which will enjoy another year of very strong growth, thanks to our robust design pipeline. And the Wi-Fi product line for new applications such as smart glasses, white-scar module, and palm-made authentication, which I mentioned in my prepared remarks, are all well positioned to contribute to growth momentum. to our kind of mainstream automotive business, now over half of our sales. It will continue to grow as well on top of a year of double-digit growth this year. In saying so, we are assuming a flattish year for global vehicle shipment, which I think is a fair assumption. imply our confidence to outperform the auto market again like we do this year and actually the last few years. Our positive outlook is driven by two factors. One, the continuous growth of number of displays per week, which has already exceeded three this year from almost now 15 years back. and two, more importantly, the fast-rising IC content per automotive display. In other words, both the number of displays and content value per display will continue the favorable trend that we have experienced over the last few years. So I will take HUD as an example. We mentioned about HUD in our preparing remarks. HUT used to be a tiny niche market that did not use LCD or image protection. Starting just a few years ago, HUT has become one of the strongest growth segments within our design wind pipeline. Now we have many, many such design winds for HUT within our line-up for mass production starting this year and over the next few years. Every new HUD in a car represents net increase in number of LCD displays per vehicle as almost all new HUDs now use LCD for image protection and IE requiring our solution. HiMAC has been leading the charge in supporting the industry's new HUD designs which require not only traditional DDIC but also very sophisticated TCAMs for what we so-called dewarping and the elimination of so-called postcard effect. Perhaps for those who are not familiar with this, dewarping is a feature in our HUD TCAM that kind of mathematically counteracts optical distortion caused by the curved surface of windshield. Okay, that's dewarping, and postcard effect is a visual artifact, so to speak, a visual artifact where the rectangular borders of the projection become kind of illuminated and visible by the driver, which is bad, right? So our T-Con can get rid of that artifact. Such advanced features add significantly to our IT content value and have become a strong mold that helps fans of competitors. and I would say L-E-D-I, which, again, we also mentioned in our previous remarks, is also a good example where we pioneered the technology concept a few years back, targeting ultra-large-sized automotive displays requiring touch functionality. We are now the exclusive supplier of the solution in the market right now, which numbers at least five chips per panel, including one T-Conc. And for higher-end designs, more than 10. Our LEDI solution is ramping nicely this year across several leading auto brands with a solid and still expanding design wind pipeline. Perhaps the last but certainly not the least important example is the higher-value display. Countable IC is an OLED display for automotive. Then we offer DDIC, TCAM, TET, and discrete touch controller, each of them having higher ASP compared to those using LCD panels because of the more complicated features required of OLED displays. And because OLED is not projected to replace LCD over time, over the next few years, for high-end mobile displays, we are working closely with several leading all the panel makers in Korea and China. With major ramp in expected starting from probably second half next year and certainly we're into 2028. So in short, again, we don't give official guidance for next year, so I'm not going to be able to quantify the growth, but we are quite positive about the prospect of our automotive business. for next year. Having said that, the industry supply will remain tight with literally all major foundries including indicated further price hikes next year. Certainly all foundries in Taiwan and to a great extent major foundries in China as well. So it's just something we will need to continue to manage like we have done so far this year. So I hope that this addresses your second question for next year's outlook. And your first question is about the CTO sales education, sales outlook for next year. Again, I will not be able to quantify that. But so qualitatively, what I can say is that for sure, CPO will see very significant growth next year. But we'll be comparing it with the low base this year. But the growth will be very significant. The extent of CPO's contribution to our overall financial next year will depend on when The end customer, Jinggao Mass Production, and obviously the rapid curve. For now, it is too early to say, but as I said earlier, even in the most conservative scenario, CPO will make a meaningful contribution to our financials, especially the bottom line. Starting 2027, with volume growth quarter over quarter, starting from this quarter, next quarter, and throughout the whole of next year. While the volume indication from end customers is actually extremely, extremely high, so we thought the main goal for us this year is still, for us and our partner, Fawcett, is still to complete the validation of our products. by key customers slash partners in terms of both quality and manufacturability. But as I said earlier, revenue contribution during this year will be limited. However, we have seen quite significant growth this quarter and another very significant growth next quarter out of engineering rounds. Again, while we don't comment on the exact timing of mass production, because the ecosystem is far too complicated for us to even speculate the exact timing of mass production, we should be able to see meaningful top and bottom line contribution from the CVO product starting in 2027. and actually even before official mass production, early shipments for engineering runs were already a positive impact on our financial starting point of concern. Now, you asked me to quantify, you know, what is meaningful. I guess meaningful means meaningful, right? Meaningful means it's not going to be like, you know, low single-digit percentage, certainly. Certainly not before the Falcon 9, right? So meaningful is something we can actually, when it happens, we can categorize it and elaborate further. It's not going to be like marginally, you know, negligible addition to our financials. But again, you know, for some early indication for next year's volume, has been provided by our customers. I think it's again this year we should focus on validation. And once we have a solid footing on that with customers, a firmer indication of mass production timetable and revenue curve, hopefully in the not too distant future we'll be able to better quantify The impact to our financials. But I guess it is fair to say that our competence level towards the success of mass production continues to rise. Our competence level continues to rise and it has surely risen to another level compared to even just a quarter. And I guess the last point I want to say is that in close collaboration with a few top customers. Fourth in us, we are focusing on what we call Gen 2 product right now, which is a high-end product supporting 6.4T bandwidth or higher. And it is a spec from time B that is positioned for the market with the biggest falling potential. and actually we will continue to make innovative designs for optics to meet the ever-rising needs of higher transmission bandwidth with the next target, next goal being 12.8G. So it's going to be a long road ahead. And again, as I said earlier, I think it's a once-in-a-lifetime opportunity for IMAX, this CP opportunity. I hope that answers your question, Tony.

Thank you, Jordan.

Thank you. Next one, team from the year, Morgan Stanley. Go ahead, please.

Thanks, Jordan. Thanks, Karen, for taking my question. So I would like to follow up on the CPO front. I think there are many peers or emerging players that want to tap into this FAU or I would say WRO market. May I ask How you would address this competition and what are Himex's key advantages over the peers? Yeah.

Thank you. Okay. Thank you. It's a good question. I think as far as we know, we are probably the only one providing optics using web-based solution or The so-called competition actually use traditional molding glass type of technology. So I think there's a key difference. And we certainly believe our technology offers a lot of advantages, but nobody has proven itself by having a big volume mass production yet. So I think it's We have to prove it by actually doing it successfully. Having said that, I think our technology together with 4C is much more mature compared to the peers, as far as we can tell. Having said that, I think, again, Our focus for now is, the mindset for now, for us, is not to worry too much about competition, but we are the enemy, we are our own enemy. For example, we focus on getting our You know, our products validated this year solidly and with customers' requests for volume, you know, try to wrap the volume successfully with good ERA and good outcome to meet their demands. And I think, so, again, we, you know, I guess also it's a policy for us not to comment about our competition. But I think for the time being, the most important thing for us is... to focus our sales efforts for successful mass production. Having said that, I think, you know, again, I said this is the market, it's a once-in-a-lifetime opportunity kind of market. The market is so big, I can't imagine a few years down the road, Himex and Hosea will be the only player, the only vendor in the market. It's just unthinkable because the market is so big and I don't think anybody can can take 100% of the market, right? So for us, I think the mindset, again, is focus on our efforts, bring successful mass production, and hope it will be one of the earliest and will be serving among the best customers. And we'll take it from there. and I think lastly, certainly we have a lot of analysis on the pros and cons of our technology versus others, but again, I'm not going to comment on that publicly. And last but not least, the market is too big. for anybody to try to take it 40%. And it's a market big enough for actually quite a few players to prosper tremendously, I think. And now the important thing is to bring it out successfully and to prove it is a real technology, is a real deal.

Go ahead, Barry. Yes, go ahead. I have a question regarding the overall CPO industry. Yes, as you guys have been working with BOCY on all these mass production preparations for so many years, I think you also in talk with other guys in the supply chain. What do you think are the biggest bottlenecks for CPO mass production right now? And you think how much time it would take for this to be conquered? Thank you.

We are very, very upstream, right? So, I mean, you probably asked the wrong person to make a comment for the so-called bottleneck throughout the entire ecosystem. What I can say, what I emphasize is that the ecosystem, admittedly, is actually quite complicated. and our focus for now is to, again, make sure we are not going to be the bottleneck, right? So we don't want to be the bottleneck ourselves, technology-wise, that's step one. And once mass production gets started, we don't want to be the bottleneck for capacity as well, and that's the goal. As to... The bottleneck of the whole ecosystem, to be honest, I really don't know, because, you know, what I can say is the engineers, the software engineers, which will have some meaningful value, right, by us at 4C over the next, by the end of this year, let's say, by the end of this year, is not just to validate our solution, but also to validate the whole ecosystem. So all technology will be used, will be taken to validate our end customers, the major, you know, GPU and CPU makers of the world. They are post-solution. And we start, they will also take their solution to their server customers. And certainly the back-end house, all offset houses do have to put the co-ords or whatnot, right? the FAU together with the TSMC scope solution and all that, right? So over the next, by the end of this year at least, I think there will be a lot of efforts in this regard. But we've been very obstinate to provide our FAU and for that our optics within FAU first. And our FAU has to will be validated in itself. Then, with good FAU, the ecosystem can start to validate the packaging, the assembly of the whole module, and then all the way down to servers and eventually probably change of algorithm because transmission methodology will be different compared to traditional Thank you, Jordan. Thank you, Tiffany. Thank you. One question from online box. How do you expect your smart glasses revenue to trend in 27? Do you expect smart glasses revenue to be large next year? We are, we are, we are talking about wide size, right? Which we, I'm not going to repeat it, we have mentioned, elaborated in our prepared remarks. Air quotes will be for the future. Of course, for now, it's too thick design wind, so that is not going to be a mass production story for next year. Wi-Fi, however, will very much be a story for next year for smart glasses. We mentioned there's a major customer which has launched its product a while back with a We saw our website serving the function of always known visual understanding of the surroundings. We actually asked the management and customer about their potential, and the response from them is they're just getting it started promotion-wise, so they have to So we are not getting one or the other conservative or pessimistic or positive feedbacks from them yet for understandable research. I think what's equally exciting is that there are quite a number of Major design-in or design-win projects in the pipeline that we certainly haven't announced because the customer is not fully ready yet. They involve, in some cases, major hyperscalers who are not known to be offering hardware devices like smartphones. as part of their business portfolio. But they are very much into smart glasses because they are all seeing smart glasses as a very good count tool for their AI models to get useful data, useful information of the actual users. So I think we are seeing very strong design momentum and which major customer they are how they use the website are not totally identical. You know, everybody is trying to be innovative and, you know, prepare remarks. What we mentioned is broadly speaking a typical like use this concept or We are always watching the surroundings on a continuous basis and then first we determine the scene that you are standing in and then we start, we then do object classification and with that you can have AI interaction. So that is a common thing. So all I can say is probably give us two or more quarters. We should be able to quantify it better. But for now, we can see the momentum. But even with the first customer, major NAM, with a very high-profile launch, we are not really getting feedback yet in terms of number of chips they need for next year or the year after. So the momentum is very strong, but I'm afraid it's too early for me to give a quantitative comment yet. Hopefully, in a few quarters of time, we will be able to provide better answers.

Yes, thank you, Jordan. There are no questions at the moment. We thank you for all your questions. Then I'll pass the call back to Mr. Jordan Wu. Please proceed. Thank you.

Thank you. As a final note, Karen Tiao, our head of IRPR, will maintain Thank you, Jordan. And ladies and gentlemen, this concludes second quarter 2026 earnings conference. You may now disconnect.

Thank you again. Goodbye.