Driven by surging demand for generative AI-related optical communications, Corning posted strong second-quarter financial results. The company declared a new phase of accelerated growth, fueled by the vast potential of AI data centers where fiber demand per GPU could increase tenfold.

Corning held its earnings call for the second quarter of 2026. Powered by the global boom in generative AI infrastructure construction, the company's Q2 sales rose 17% year-over-year to $4.74 billion. Earnings per share (EPS) surged 30% to $0.78, with both gross margin and operating margin expanding significantly.

Corning expects third-quarter sales to be between $4.9 billion and $5.0 billion, with core EPS growing approximately 28% year-over-year. CEO and President Wendell P. Weeks stated on the call: "We are entering a new phase of accelerated growth."

It is important to note that the top end of Corning's Q3 sales guidance only matched expectations, and the implied sequential growth rate dropped sharply from 9% to 3-6%. The company also cautioned that the actual pace of optical scaling adoption remains one of the biggest variables in its future growth forecast. As of this writing, Corning shares were down 18% in after-hours trading.

AI Ignites Optical Communications: 'Orders Are Accelerating'

Among Corning's business segments, the optical communications division was undoubtedly the standout performer in the quarter and the absolute core supporting future high growth. In Q2, segment sales surged 32% year-over-year, surpassing the $2 billion mark. Net profit soared 77% to $438 million, a record high.

This explosive growth was directly fueled by massive global tech investment in AI computing infrastructure. Weeks emphasized on the call: "Our generative AI products and enterprise network demand remain strong, and orders are accelerating."

He revealed that since the launch of the company's "Springboard Plan," enterprise sales have more than tripled. In Q2 alone, they grew 65% year-over-year to $1.27 billion, with "generative AI product sales nearly doubling."

To secure key customers, Corning has forged long-term, large-scale commercial agreements with giants like Apple, Meta, Nvidia, and Amazon to share the risks and costs of capacity expansion.

AI Factory Computing Power Surge: Fiber Demand Per GPU Could Increase Tenfold

Addressing the market's central question of how much growth AI will bring, management provided a compelling calculation framework. Weeks stated: "At the most basic level, assuming no network changes, we will grow with GPU growth."

However, network architecture is undergoing radical changes that will allow Corning's growth to outpace GPU growth itself. He highlighted three key drivers: cluster size growth, bandwidth growth, and scale-up.

In the scale-up domain, networks are traditionally 100% copper, but with the introduction of new architectures like Nvidia's Vera Rubin Ultra, optics are beginning to penetrate. Weeks calculated that if a scale-up network transitions entirely from copper to optical, combined with scale-out needs, "the total fiber content per GPU reaches 160 fibers, which is 10 times the fiber count of the current scale-out network."

While the reality may be a hybrid system, Weeks clearly stated this represents "a very large opportunity for us." By 2028, optical content per GPU in the enterprise market could increase by 1.3 to 1.5 times, and by 2030, this figure could become even higher.

CFO Edward A. Schlesinger noted that when developing the company's high-confidence plan, they risk-adjusted for factors including the macro environment, policy changes, technology adoption timelines, and customer onboarding speeds. One of the most significant adjustments, he said, is the adoption pace of network scale-up. "The adoption of optical scale-up in AI factories is a major technological change. The overall opportunity is enormous, but timing is challenging to predict."

A New Business 'Inside the Box': Targeting a $10 Billion Photonics Market

Beyond traditional external fiber connections, AI's relentless pursuit of higher compute density and lower latency is opening a new door for Corning: the 'inside-the-box' passive photonics market.

As the industry shifts from pluggable modules to co-packaged optics (CPO) and near-packaged optics (NPO), light generation and transmission are moving to the silicon photonic optical engine. Weeks pointed out that this "creates an opportunity for Corning's passive photonics to manage light. Previously, we had no content 'inside the box.'"

For this incremental market, Corning provided an optimistic outlook: "Based on our assumptions and discussions with customers, we believe there is an opportunity to build a $10 billion market access platform by 2030."

'Springboard Plan' Accelerating: Targeting $40 Billion in Revenue by 2030

Based on the strong recovery across business lines and the significant AI dividend, Corning management reiterated and refined its "20-30-40 Springboard Plan."

CFO Edward A. Schlesinger outlined the internal plan: "Increase our annualized sales run rate to $20 billion by the end of 2026, $30 billion by the end of 2028, and $40 billion by the end of 2030."

Even after risk adjustments for macroeconomics, policy, and technology adoption rates, Corning's high-confidence plan still projects sales of $35 billion by the end of 2030, effectively doubling the company's size. Financially, Corning expects a compound annual growth rate (CAGR) of 19% in sales from Q4 2026 to Q4 2030. Schlesinger reassured the market: "You can expect us to continue to maintain operating margins of 20% or more, even as we continue to invest to capture all growth opportunities."

Full Transcript of Corning's Q2 2026 Earnings Call

Wendell P. Weeks, CEO and President: Good morning, everyone. Today, we announced excellent second-quarter results, demonstrating our progress on the recently upgraded Springboard Plan. For those of you who have been with us on this journey, you'll recall we launched the Springboard Plan in Q4 2023 with an annualized sales run rate of $13 billion.

Over the past two and a half years, we have significantly increased sales and successfully transformed the company's financial profile. Our plan is to increase the annualized sales run rate to $20 billion by the end of 2026, $30 billion by the end of 2028, and $40 billion by the end of 2030. We are entering a new phase of accelerated growth. We expect a sales CAGR of 19% from Q4 2026 to Q4 2030, with earnings growing faster than sales, a significant improvement in return on invested capital, and substantially increased free cash flow.

Against this backdrop, let me detail our Q2 performance. Year-over-year, sales grew 17% to $4.74 billion. EPS grew 30% to $0.78. Gross margin expanded 120 basis points to 39.6%. Operating margin expanded 190 basis points to 20.9%. ROIC expanded 180 basis points to 14.9%, and free cash flow increased to $1.42 billion.

Our performance was primarily driven by the optical communications segment, where sales grew 32% year-over-year to over $2 billion, and net income grew 77% to $438 million. Our generative AI products and enterprise network demand remain strong, and orders are accelerating. Since the Springboard Plan began, our enterprise sales have more than tripled. In Q2, we grew 65% year-over-year to $1.27 billion, with generative AI product sales nearly doubling. Remember, these are all scale-out. Our results do not yet include scale-up or photonics. Overall, we face a significant opportunity in optical communications, which I will detail shortly.

Turning to the solar business. Our sales grew 90% year-over-year, and we completed an extended maintenance shutdown and equipment upgrade at our solar wafer factory. We expect improved sales and profitability in Q3. During the quarter, key customers continued to choose our latest innovations and support the expansion of our manufacturing platform to accelerate their and our growth plans. Reflecting on our progress, last year Apple expanded our long-term partnership, committing to produce 100% of iPhone and Apple Watch cover glass at our Kentucky facility. In Q1, Corning and Meta announced a multi-year agreement valued at up to $6 billion, leveraging our latest innovations in fiber, cable, and connectivity solutions to support Meta's applications, technology, and AI vision.

In May, Nvidia announced a multi-year commercial and technology partnership with Corning to significantly expand the U.S. domestic manufacturing of advanced optical connectivity solutions powering next-generation AI infrastructure.

In June, Amazon announced a multi-billion dollar agreement with Corning to provide fiber, cable, and connectivity solutions supporting Amazon's expanding data center infrastructure across the U.S. These deep customer partnerships support the extraordinary growth outlined in the upgraded Springboard Plan we shared at our May investor event.

Now, as you know, we provided a lot of exciting news and details at that event. If you missed it, I encourage you to visit our website for the presentation. This morning, I'll briefly recap the key takeaways. Our internal Springboard Plan is to increase the annualized sales run rate to $20 billion by the end of 2026, $30 billion by the end of 2028, and $40 billion by the end of 2030. For simplicity, we refer to this as our 20-30-40 Springboard Plan.

As a reminder, our internal plan is the output of our strategic planning process with each market access platform. These are our actual business plans. We set goals and compensation against these plans. When our business units submit plans to the company, they consider various probabilistic outcomes. They try to account for known unknowns. The goal of the business plan is a 70% confidence interval, meaning they have a 70% probability of achieving sales greater than or equal to that number. We then translate the internal plan into an enterprise-level risk-adjusted high-confidence plan for investors, which Ed will review in a moment.

I will share some key assumptions from our internal plan. For 2027 to 2030, we use a forward exchange rate of 150 yen to the dollar to account for the weaker yen. We plan for flat TV, IT, and smartphone end markets and consider the impact of memory price increases. We plan for a decline in internal combustion engine demand, offset by increased Corning automotive glass content. We also plan to capture a larger solar opportunity with an upgraded sales outlook. We incorporate new innovations and form factors for Gorilla Glass. We see accelerating growth in fiber-to-the-home and carrier data center interconnect.

With this context, let's look at the company's growth across segments. First, I will show a chart of our total revenue base. We are entering an accelerated growth phase. In the first phase of the Springboard Plan (through Q4 2026), we expect to achieve an attractive 15% sales CAGR, along with a significantly improved financial profile.

Comparing Q2 2026 to the start of the Springboard Plan, our EPS has doubled, operating margin has expanded 460 basis points, and ROIC has expanded 610 basis points. Overall, we have established an excellent foundation for future high-margin growth. From this strong base, entering 2027, we expect our growth rate to accelerate to a 19% CAGR, a 400 basis point increase.

We expect growth in consumer electronics, solar, carrier, automotive, and life sciences. Overall, we plan for these markets to grow at a mid-single-digit CAGR. We plan to continue the Springboard Plan approach, providing investors with deeper updates and insights as individual markets reach important milestones.

At our May event, we just reached such milestones in the enterprise market and photonics, so these were the focus of our presentation. We are working in a rapidly evolving field where multiple views exist on future AI network architectures. Since our May release, our fundamental view has not changed. So, I want to reiterate the key takeaways.

Starting with the enterprise market, we have the ability to grow faster than GPU growth, driven by the technical factors that increase optics in data centers. At the most basic level, assuming no network changes, we will grow with GPU growth.

Now, everyone has their own views on GPU growth. The point we want to reiterate today is that several potential network changes provide us with an opportunity to outpace GPU growth in the enterprise market. We will describe the technical drivers, the logic, and their respective impacts.

The first driver is cluster size growth. The logic is that cluster sizes exceeding 130,000 GPUs will require a third optical layer. As clusters grow, this is favorable for our content opportunity.

As shown in the chart, once a cluster exceeds 130,000 GPUs, it exceeds the network scale capability of a two-layer network using 512 Radix switches. This necessitates adding a third layer. Essentially, three layers divided by two layers means a 50% increase in content per GPU for hyperscale clusters. These large clusters are the fastest-growing part of AI factories. Therefore, cluster size growth is a tailwind for Corning relative to GPU growth.

Next, the second driver. The second driver is bandwidth growth. Historically, GPU and ASIC bandwidth roughly doubles every two years. We connect them through a combination of channel rate and channel count. Typically, this has a neutral to positive impact based on the SerDes cycle. We increase bandwidth either by increasing the channel rate or SerDes (neutral impact on fiber content) or by increasing the channel count (potentially positive impact on fiber content).

You can see that when we transitioned from Hopper to Blackwell, the SerDes remained at 100G, but bandwidth needed to double, so it required us to increase fiber from 8 to 16, doubling the content. As we move into the Rubin era of GPU architecture, we see the SerDes jump to 200G. Therefore, we can keep the channel count consistent, resulting in a neutral impact on fiber content. Feynman likely won't be a major system until 2029-2030. There is still much we don't know about it. But if it follows past patterns and stays at 200G, then the channel count would double, bandwidth would double, and fiber would double again. Alternatively, if 400G SerDes is available, fiber content would be neutral or unchanged.

Similarly, other optical schemes like BiDi and WDM can improve fiber efficiency, potentially reducing the need for more fiber per GPU. This is not yet settled. We will know more in about a year, but the main takeaway is that bandwidth is neutral or positive for us. In our 20-30-40 Springboard Plan, we assume bandwidth has a neutral impact on fiber count per GPU.

The third driver is scale-up. Currently, this is 100% copper. But optics are beginning to penetrate scale-up networks. This involves an entirely new optical network. While the timing of adoption and penetration is difficult to predict, the magnitude of the optical content opportunity is substantial.

First, consider what has been announced regarding optical scale-up. Recently, Nvidia announced the Vera Rubin Ultra configuration, which will scale up to 576 GPUs across eight independent racks. Each rack will have 72 Rubin Ultra GPUs, interconnected via copper, and then via direct optical connections for inter-rack scaling. This is a step towards optics, effectively a hybrid system approach to scale-up. Optics are now playing a role.

The specific percentage of optical ports has not been publicly disclosed. What has been disclosed is 1.6 terabits per second of scale-out bandwidth and 14.4 terabits per second of scale-up bandwidth per GPU. So let's examine the range of this opportunity.

At the low end, we can assume 100% of the scale-up network will be copper, as it is today. This means the same opportunity as today: no fiber in scale-up, and 16 fibers per GPU in scale-out.

Now let's compare this to an all-optical scale-up system. We take the 14.4 Tbps per GPU for scale-up and 1.6 Tbps per GPU for scale-out, divided by 200G SerDes. This translates to 72 channels and 8 channels, respectively, with each channel requiring two fibers. This means 144 fibers are needed to support scale-up bandwidth and 16 fibers to support scale-out bandwidth. When we combine these requirements, the total fiber content per GPU reaches 160 fibers, which is 10 times the fiber count of the current scale-out network. What we know for sure is that neither of these scenarios is the just-announced hybrid system. The reality will be somewhere in between.

To accurately understand this opportunity, we need both the percentage of optical ports in that product and how much these new hybrid optical scale-up nodes will penetrate AI factories. Unfortunately, I cannot share the former as it is confidential. And the answer to the second question—how successful these solutions will be—no one knows for sure. But this is clearly a very large opportunity for us. It is a topic of much technical debate, and you can form your own view by talking to experts.

When I synthesize all these technical drivers, we calculate that by 2028, the optical content requirement per GPU in our enterprise market will increase by 1.3 to 1.5 times. By 2030, this number could potentially be higher. Most of this growth is driven by the rapid growth of the scale-up opportunity, which leads us to our next incremental opportunity 'inside the box.'

Scale-up also dramatically increases our opportunity in the new photonics market, serving a new class of customers. We are introducing optics 'inside the box' for next-generation technologies like co-packaged optics and near-packaged optics. While these technologies may start with scale-out, it is clear that scale-up significantly increases the magnitude of the opportunity. Optical scale-up is a new technology that may see an exponential adoption curve, making timing challenging to predict. Based on our assumptions and discussions with customers, we believe there is an opportunity to build a $10 billion market access platform by 2030.

Essentially, the new 'inside-the-box' optical function creates an opportunity for Corning's passive photonics to manage light. Previously, we had no content 'inside the box.' What is happening is that customers are seeking to transition from pluggable solutions to CPO and NPO solutions due to the potential for improvements in latency, panel density, power consumption, and reliability.

So, as you can see in this chart, the generation, modulation, and transmission of the encoded optical signal have now moved to the silicon photonic optical engine 'inside the box.' All the parts highlighted in yellow represent potential Corning content that did not exist 'inside the box' before. This provides Corning with the opportunity to supply the passive photonics required to move and manage light.

Okay, we've covered a lot of information. Clearly, this is a greatly simplified version of the entire presentation we shared at the May investor event. Again, if you haven't seen it, I encourage you to watch the full presentation on our website.

Before I hand the call over to Ed, let me summarize the key takeaways I want to leave with you today. We had an excellent Q2 performance demonstrating our progress on the 20-30-40 Springboard Plan. Our plan is to increase the annualized sales run rate to $20 billion by the end of 2026, $30 billion by the end of 2028, and $40 billion by the end of 2030. We are entering a new phase of accelerated growth. We expect a sales CAGR of 19% from Q4 2026 to Q4 2030, with earnings growing faster than sales, a significant improvement in ROIC, and substantially increased free cash flow.

We expect growth across the company, with particularly significant opportunities in the enterprise network and photonics markets. In the enterprise market, we expect strong growth as AI data center clusters increase in scale-out and optical scale-up is adopted. In our new photonics market, we plan to build a $10 billion revenue stream by 2030.

We continue to deepen our relationships with industry leaders, as evidenced by our recent partnerships with Amazon and Nvidia. These long-term partnerships support our extraordinary growth opportunity. We are clearly in a very exciting period for the company. We plan to continue the Springboard Plan approach, providing investors with deeper updates and insights as individual markets reach important milestones.

I look forward to updating you on our progress as we work towards doubling the company's size over the next few years. We are very excited to have you with us on this journey.

Now, I'll turn the call over to Ed. Ed?

Edward A. Schlesinger, CFO: Thank you, Wendell. Good morning, everyone. I am very pleased with our strong second-quarter results. We delivered another quarter of double-digit year-over-year sales growth while continuing to improve our financial profile. Year-over-year in Q2, sales grew 17% to $4.74 billion, and EPS grew 30% to $0.78, both above our guidance range. Operating margin grew 190 basis points to 20.9%. ROIC improved 180 basis points to 14.9%, and we generated $1.42 billion in free cash flow.

Turning to the business segments, starting with optical communications, sales were $2.07 billion, up 32% year-over-year. Net income was $438 million, up 77% year-over-year. The segment achieved record profitability in Q2, with a net income-to-sales ratio of 21%. Enterprise market sales grew 65% year-over-year, driven by continued strong demand for our generative AI innovations, with orders accelerating. The AI data center-related portion of the enterprise market nearly doubled in the quarter.

Carrier market sales in Q2 were up 1% year-over-year. Over the long term, in the carrier market, we expect mid-single-digit sales growth, driven by fiber-to-the-home deployments and data center interconnect.

Across the entire optical communications segment, we continue to expand and strengthen strategic agreements with our key customer base, solidifying Corning's position as a critical supplier for next-generation AI and broadband infrastructure. Turning to the glass innovation business. Q2 sales were $1.46 billion, up $20 million or 1% year-over-year, driven by growth in display glass sales. Net income was $354 million, up 9% year-over-year. Now, we've received many questions about the impact of memory prices. For the full year, we expect memory prices to impact the handheld device market, with volumes declining in the mid-teens percentage range. Despite these headwinds, we expect Gorilla Glass sales to outperform the end market, driven by strong demand for our innovations and our position in the premium segment of the market. We saw this dynamic in the first half of the year as well.

Even in a market downturn, our 'More Corning' strategy of increasing content per device through products like glass ceramics or foldable displays is having a positive impact. In the display market, the impact of memory prices is expected to be less significant. In fact, as component costs rise, TV brands and panel makers are shifting to higher-priced, larger-sized TVs, which benefits our strong position in Gen 10.5 glass.

In the advanced optics area, we expect strong demand for advanced memory to support long-term demand for our solutions from chipmakers and semiconductor equipment suppliers.

In our automotive business segment, Q2 sales were $471 million, up 2% year-over-year. Net income was $82 million, up 4% year-over-year. Our sales to the automotive market grew 2%, driven by more Corning content, outperforming the global automotive market (which was down 2%).

Diesel business sales grew 3% year-over-year and 13% sequentially, driven by improved North American Class 8 truck orders. Looking ahead, we remain focused on our 'More Corning Content' strategy. We expect the underlying long-term trends favorable to Corning to remain intact, driving larger sizes and higher resolution for in-vehicle displays, as well as the adoption of new emissions control products.

In the solar business, Q2 sales were $438 million, up $207 million or 90% year-over-year. The segment reported a net loss of $7 million. In Q2, as expected, our solar wafer factory underwent an extended maintenance shutdown, transitioning to a permanent power system while repairing and upgrading production equipment, resulting in $30 million in additional expenses compared to Q1. Customer demand was strong across the market, and we expect sales and profitability to improve starting in Q3.

Overall, from a business perspective, we continue to secure long-term customer commitments for polysilicon, wafers, and components. Market preference for U.S.-manufactured solar products continues to strengthen, supported by ongoing trade and tax policy developments and other government initiatives to advance domestic manufacturing.

Corning remains well-positioned in this area as the only U.S. manufacturer of polysilicon and wafers. We remain firmly on track to build this solar business into a $3 billion revenue stream with profitability above the company average.

Life sciences and emerging growth businesses saw sales increase 8% sequentially, driven by strong performance in the life sciences research business, and net income increased 13% sequentially.

Turning to our outlook. For Q3, we expect sales to be up approximately 16% year-over-year, in the range of $4.9 billion to $5.0 billion, and core EPS to be up approximately 28% year-over-year, in the range of $0.85 to $0.89. In our solar business, we expect sales and profitability to improve starting in Q3.

Regarding capital expenditure, we expect investment rates to increase in Q3 and Q4, with full-year capital spending of approximately $2 billion to support the compelling growth plans in optical communications that Wendell just described. For the full year, we remain on track to deliver a significant year-over-year increase in free cash flow, while continuing to invest in our growth opportunities alongside our customers.

Now, before we move to Q&A, let me return to our Springboard Plan. I will begin by reiterating how we translate our internal plan into a high-confidence plan to help you with your investment decisions.

Our internal Springboard Plan is to increase the annualized sales run rate to $20 billion by the end of 2026, $30 billion by the end of 2028, and $40 billion by the end of 2030, with earnings growing faster than sales.

Our high-confidence plan is to increase sales to an annualized run rate of $27 billion by the end of 2028 and $35 billion by the end of 2030. Interestingly, in either case, we expect to double the size of the company by the end of 2030. If we achieve our internal plan, we will double the sales run rate between Q4 2026 and Q4 2030. Our high-confidence plan will double the run rate between Q4 2025 and Q4 2030.

As a reminder of how these plans work, our internal plan is the output of our strategic planning process with each market access platform. These are the actual business plans. We set goals and compensation against these plans.

To derive our high-confidence plan, we apply further risk adjustments to the internal plan. At the enterprise level, we attempt to probabilistically adjust for factors including macroeconomic slowdowns, government policy changes, the timing of multiple long-term trends, and the adoption rate of our related innovations.

One of the most significant areas we are adjusting is the timing of network scale-up. This impacts both the enterprise and photonics areas. The adoption of optical scale-up in AI factories is a major technological change. The overall magnitude of the opportunity is enormous, but timing is challenging to predict. We will gain more understanding of this over time. As we have done throughout the Springboard Plan, we will provide updates and milestones to help investors track the progress of both plans.

Overall, we have established an excellent foundation for future high-margin growth. If we compare the Q2 2026 results announced today to Q4 2023 when the Springboard Plan launched, our sales have grown 45%, operating margin has improved 460 basis points, EPS has grown 100%, and ROIC has expanded 610 basis points. So, we are operating from a very strong financial base, and we expect the financial profile to improve further from here.

Our plan is to achieve a 19% sales CAGR from Q4 2026 to Q4 2030. You can expect us to continue to maintain operating margins of 20% or more, even as we continue to invest to capture all growth opportunities. We will come back later this year to update you on our view of operating margins. Our EPS growth has been outpacing sales growth, and we expect this to continue. We have already significantly improved ROIC to approximately 15%, and we expect to continue to improve ROIC to near 20% over the planning cycle. Most importantly, we expect significant growth in free cash flow. Typically, when we make organic investments, we put significant capital to work upfront, meaning we take on risk before we see the revenue and free cash flow. As part of the Springboard Plan, we are deepening key customer relationships through long-term agreements to share the risks and costs of capacity expansion more appropriately with our customers. The result will be attractive, and we expect free cash flow to grow even as we invest to capture higher sales.

Overall, we have outlined a compelling new plan to further enhance our financial profile. The performance and progress on the key Q2 milestone we just shared demonstrates we are off to a strong start. We look forward to keeping you updated on the progress of this significant value creation opportunity. With that, I'll turn the call back to Chris for Q&A.

Christopher Keenan, Investor Relations: Thank you, Ed. Operator, we are ready for the first question.

Question-and-Answer Session

Operator: (Operator Instructions) One moment, please. Our first question comes from Asiya Merched from Citi. Please go ahead.

Asiya Merched, Analyst: Great. Good morning. Very nice results. If I could just dig in a little bit -- Wendell and Ed, on the growth outlook for the second half, particularly Q4, given you gave guidance for Q3, optical capacity is ramping, the solar plant migration you mentioned was completed in Q2. Why is Q4 -- assuming an exit run rate of $5 billion -- why is Q4 growth very limited? Is there anything else in other markets we need to consider? Thank you.

Edward A. Schlesinger, CFO: Thanks, Asiya. First, as we shared, we expect sales to continue to grow significantly from the current run rate over the next planning cycle, going from a $20 billion run rate to $30 billion to $40 billion over the next four years or so. In Q3, our guidance implies we might be about a quarter ahead of schedule in reaching the $20 billion run rate by the end of the year, and then we expect to build on that from there.

I think the most important thing is that our year-over-year growth rates have been in the mid-teens, around 15%. We are starting to see the growth rate accelerate, and we actually expect our growth rate to accelerate further to near 20% from the end of this year through the end of the planning cycle. Our CAGR is 19%. So, I don't think we are signaling anything specific about any other market. I think the majority of the growth will come from the enterprise market and photonics, as we have shared, and of course, we expect the solar business to grow from current levels.

Wendell P. Weeks, CEO and President: Simply put, in May we said we would reach the $20 billion run rate by Q4, and we are growing faster. So it looks like we might hit it a quarter early, and we didn't want to upgrade our Springboard Plan again right after updating it in May. So, when we next update and provide Q4 guidance, we will. In the meantime, no implied message.

Asiya Merched, Analyst: Great. Thank you very much. If I could ask one more on optical margins. It looks like there should be good growth in the second half. You have the capacity, you expect to ramp to meet demand, and orders seem to be accelerating, especially in AI. It would be helpful if you could help us think about how optical net margins should be thought about. Thank you.

Edward A. Schlesinger, CFO: Yes. Thanks, Asiya. Last quarter we were at 20%. This quarter, we exceeded 20%, a significant improvement from where we were at the start of the Springboard Plan. I think we will continue to see good growth in optics, and we certainly see margin expansion as well.

Christopher Keenan, Investor Relations: Thanks, Asiya. Next question?

Operator: From UBS, Josh Spector. Please go ahead.

Josh Spector, Analyst: Yes. Hi. Good morning. I wanted to ask about the investor day update, glad you recapped it. But to be clear, has anything changed compared to two months ago? Particularly in photonics and scale-up, there is a lot of industry discussion about supply chain readiness and potential delays. I mean, you seem to be reaffirming that photonics could be a $1 billion-plus business next year. Do you still have that visibility? Would you describe things differently today versus two months ago?

Wendell P. Weeks, CEO and President: So what you are hearing is that we have not seen any change in our fundamental beliefs that we just shared in May. Yes, there is a lot of speculation in the industry about exactly when things will start. That's fair. You have to understand that when we provided that Springboard Plan, we were adjusting for different probabilities of different timing and different content opportunities.

What we have done is made those judgments based on understanding where different product mixes and timing might go. So we haven't seen a real change. Inside the real ecosystem, it is far less dramatic than it looks from the outside. We are doing solid work to build a $10 billion new photonics platform and to bring an important new technology node to the overall AI, which is scale-up, where our biggest opportunity lies.

And inside the ecosystem, the change is much less than what is speculated outside, because we all know internally that there are various trade-offs we can make to still achieve scale-up based on changes in other component suppliers and chip dynamics.

Does that make sense, Josh?

Josh Spector, Analyst: Yes. No, it does. Thank you.

Wendell P. Weeks, CEO and President: And I think we didn't give any guidance on photonics for next year, just to be clear, right? We gave the long-term $10 billion target here, gave the growth chart, but we haven't provided you with specific numbers for photonics yet. So the $1 billion photonics analysis you did, I understand. I understand how you got those numbers. We will update each quarter, and we gave you the long-term photonics plan we laid out in the 20-30-40 plan.

Josh Spector, Analyst: Okay. Thanks. Good luck.

Christopher Keenan, Investor Relations: Thanks, Josh. Next question.

Operator: Thank you. From Wolfe Research, George Notter. Please go ahead.

George Notter, Analyst: Hi, everyone. Thank you very much. I'm just curious about the mix of the optical business that is covered by long-term agreements. I'm wondering what it looks like now. Wondering what it might look like a year from now.

And then I guess I would assume you are taking a more aggressive pricing posture on the optical business for customers not under long-term agreements. I'm curious what that looks like now, what price increases you are taking? Any perspective would be appreciated. Thank you.

Wendell P. Weeks, CEO and President: Thanks for the question, George. You can expect that we will continue to increase the number of long-term agreements, because all of our significant capacity expansions are supported by these agreements, and we seek to appropriately share risk and reward with customers in response to the strong growth demand for our capabilities. So it will be a steadily increasing cadence.

Sometimes other customers want to be public, and we announce it. Sometimes not, but the work continues, and we see more and more customers wanting to get more from us, and they are willing to commit to support whatever investments we need to make. That's the first part of your question. Did I answer that satisfactorily, George, before I continue --

George Notter, Analyst: Yes, great.

Wendell P. Weeks, CEO and President: Okay.

George Notter, Analyst: Yes, I was just curious if it's a small part of the overall optical business or a major part --

Wendell P. Weeks, CEO and President: It will be the vast majority of our optical business, and when I said any significant capacity expansion will be supported by these agreements that appropriately share risk and reward. Given our growth rate, just mathematically, this will become the overwhelming portion of our overall optical business. Does that make sense?

George Notter, Analyst: Yes. It does.

Wendell P. Weeks, CEO and President: Okay.

George Notter, Analyst: And then pricing?

Wendell P. Weeks, CEO and President: Now let's go to pricing. I didn't forget the second part of your question, George.

George Notter, Analyst: Of course. Thank you.

Wendell P. Weeks, CEO and President: Let's start with what you see in the results, and then how we do it, right? You just saw -- Ed just showed you optical sales up about 30%, net income up nearly 80%. So you are seeing a significant improvement in profitability, right?

Now, how we do that is not just by raising the price of bare fiber. Prices are higher, but that's not really how we drive the profitability improvement. We create value by providing innovations in fiber, cable connectivity that dramatically reduce our customers' costs, or improve their deployment speed, or improve the reliability of their network. When we create value for our customers through innovation, we are able to capture a portion of that innovation value. That is what you are seeing in the significantly improved profitability.

I expect this to continue. Your question also implicitly asked about supply-demand dynamics. We continue to have an enviable situation: if we can produce more, we can sell more. Especially if we can produce more of our most innovative products, we see demand for these new high-density product lines that we started developing a long time ago only growing, and their adoption is definitely accelerating.

Christopher Keenan, Investor Relations: Thanks, George. Next question.

Operator: Thank you. From Bank of America Merrill Lynch, Vamsi Mohan. Please go ahead.

Vamsi Mohan, Analyst: Yes, thank you very much. I was wondering if you could elaborate a bit more on the components of the Q3 guidance. At the consolidated level, with the midpoint, your Q3 guidance implies some deceleration, and I hear your positive tone about re-acceleration over the longer term. Q2 was obviously supported by very strong optical enterprise strength.

So the question is, do you expect that strength to persist? I think Wendell mentioned that generative AI-related sales might have doubled in Q2. Is that a trend you expect to continue into Q3, Q4? Or is there something else causing a slight gap in Q3?

Edward A. Schlesinger, CFO: Yes, hey, Vamsi. Thanks for the question. So our Q3 guidance is not meant to imply any deceleration in growth. I think it's similar -- it's meant to imply a similar year-over-year growth rate to Q2. And then as we have shared for the long term, we expect the growth rate to actually accelerate. So that's how you should think about it overall. And of course, enterprise growth is an important part of the Q3 picture.

Vamsi Mohan, Analyst: Anything else to add, is there any end market that is below seasonal for any particular reason in Q3? And maybe I'll also ask this. Your enterprise sales grew about $300 million sequentially from Q1 to Q2, but you didn't see much operating leverage in your net margins. What was offsetting the leverage on that much sales growth? Thank you very much.

Edward A. Schlesinger, CFO: Yes. Regarding the first part of your question, I would say, as I shared in my prepared remarks, the area that might not be following normal seasonality is probably the memory-affected areas, like the handheld device market overall will be significantly weaker in the second half. We will outperform that market because we sell more content per device, but it will certainly have an impact. The automotive market remains relatively subdued, but I don't think that has a huge seasonal impact, but these are similar dynamics to what you saw in Q2. And then on the margin side, I think we continue -- I think Wendell's description of the specific optical margins is very compelling from our perspective, and we expect those margins to continue to improve as we grow and sell more new innovative products where we can capture more value. Again, as we grow, Corning's overall operating margins are also expanding.

Christopher Keenan, Investor Relations: Thanks, Vamsi. Next question.

Operator: Our next question comes from Joseph Cardoso from JPMorgan. Please go ahead.

Joseph Cardoso, Analyst: Hey, good morning, thanks for the question. I know there are some assumptions here, but when I do some rough math just based on the first half and the Q3 guidance, it seems to imply a nice step-up in gross margin and operating margin, and a nice improvement in incremental margins. First, is that directional assessment correct?

And second, if so, what are the various pieces driving the step-up in Q3? And how should we think about sustainability, especially in the context of the prepared remarks about maintaining the 20%+ operating margin target? Just want to dig into that a bit more. Thank you.

Edward A. Schlesinger, CFO: Yes, Joe, I think your assessment is correct. One thing I would point out is that we expect the solar business's profitability and earnings to improve from Q2 to Q3, as we have gotten through the cycle of improving manufacturing performance. That is certainly a driver.

As we have shared more broadly, even dating back to the May investor relations event, we expect to meet or exceed the 20% operating margin target we set a few years ago. We haven't set a new target yet. We will come back later this year. We'll talk a bit about that. But I think the thing to consider is that we are still ramping the solar business, so it's not fully complete. Therefore, profitability will continue to improve. We want to see that. We want to make more progress there.

And then we have to build a very large photonics business, as we see network scale-up and photonics ramping, and we want to continue to see that as well. We feel good about the profitability, but before we change the target, we just want to have a bit more experience under our belt.

Joseph Cardoso, Analyst: No, thank you. I appreciate it.

Christopher Keenan, Investor Relations: Thanks, Joe. Next question.

Operator: Thank you. From Morgan Stanley, Meta Marshall. Please go ahead.

Meta Marshall, Analyst: Great. Thanks. Maybe I wanted to ask about the carrier business, whether this reflects a challenging comparison, like some new customer growth last year? Or is it a reallocation of enterprise share? Just trying to understand the performance this quarter versus the signal that you still say mid-single-digit growth? And then as a follow-up, there has been a lot of discussion about making new incremental investments to strengthen certain routes or add new incremental long-haul capacity. Any comments on participation in that? Thanks.

Wendell P. Weeks, CEO and President: So, it's just -- it's purely a matter of customer timing, what you saw in the quarter. If you look at the first half of 2026 versus the first half of 2025, carrier sales grew about 17%, about $250 million. It's just the way the carrier business works, right? What project is happening when, and the timing of our work with that particular customer. But we don't see any lack of demand growth on the carrier side, whether it's in DCI, as you mentioned, or in fiber-to-the-home. So it will continue to be a nice growth rate.

Does that answer your question, or do you have a follow-up?

Meta Marshall, Analyst: Yes. And then just -- I mean, there has been discussion about Verizon and Google, just incremental projects coming online, do you feel like you are able to participate in those additional growth opportunities?

Wendell P. Weeks, CEO and President: I'll put this in the same category as our comments on the enterprise market just now. So our -- customers do want our new product portfolio, as long as we can produce it. So any and all of these new opportunities almost always come to us first. Then -- so we have a considerable amount of demand. What we need to do is make sure we are positioned to continue to build these great customer relationships that will last for decades, giving us the opportunity to innovate and create value, and give our investors an opportunity to rely on a long-term annuity. That's what drives our customers' choices.

Meta Marshall, Analyst: Great. Thank you very much.

Christopher Keenan, Investor Relations: Thanks, Meta. Last question.

Operator: Thank you. From Susquehanna, Mehdi Hosseini. Please go ahead.

Mehdi Hosseini, Analyst: Thanks for taking my question. Most of the good ones have been asked. So I have two follow-ups. Starting with Wendell, there is an increasing push to bring semiconductor manufacturing to the U.S. We have a prime wafer maker, ex-NEMC. In this context, I'm wondering why not allocate a larger percentage of polysilicon to electronic grade, where there is a clear path to improve profitability, and it's also in line with the U.S. manufacturing ethos. I have a follow-up.

Wendell P. Weeks, CEO and President: I think, to your point, we expect that more of our product mix will -- we will increase production of polysilicon for semiconductors, especially the highest grade semiconductor. That's something we will continue. As a percentage of our overall revenue mix, solar is much larger in volume, but we will -- I think your suggestion is a good one, and you should expect that we will continue to increase our participation in the highest grade semiconductor polysilicon.

Mehdi Hosseini, Analyst: Okay. That should help the solar segment's profitability.

Wendell P. Weeks, CEO and President: Yes, it should.

Mehdi Hosseini, Analyst: Second question -- thank you very much for all the details on optics, you provided a lot of insights that can be applied across the supply chain. What I wanted to get from you is to what extent scale-up and the fiber array unit are already included in the Springboard Plan? You went into a lot of detail, but I'm a bit confused about how much of that is already baked into your Springboard Plan?

Wendell P. Weeks, CEO and President: I completely understand where the confusion might come from. Because we deliver fiber, right, to the panel in the enterprise market, and with scale-up, this has the potential to increase dramatically. And then what we call photonics is the content inside the panel, which includes the FAU harness or fiber array unit harness that you mentioned.

So that's what we are putting in that photonics number, that photonics market, you saw $10 billion. FAU is part of it. There is more in that photonics market. The best way to understand it is if you look at the chart I showed, all the parts in yellow, that's the potential Corning content, which will give you a good idea of the various different products in it.

(Multiple speakers) Go ahead. Sorry, please continue.

Mehdi Hosseini, Analyst: Sorry to interrupt. So the $10 billion baseline assumption does include some of the opportunities highlighted in yellow on slide 26, but there could be upside, or it's just based on assumptions. We don't know the slope of adoption.

Wendell P. Weeks, CEO and President: Yes, it depends heavily on -- the biggest driver will be, if you ask the right question, what percentage of the ports are optical, right? And then how successful is optical scale-up in AI factories? And this dynamic, you are seeing that you are going to have two parts of scale-up, which is why we are watching it so closely, that both our enterprise product and our 'inside the box' or photonics product will increase significantly.

That's why I laid out those exact technical drivers, which you can follow as an investor and talk to others to form a view, because the answers to these questions determine how much faster we grow than GPU growth.

Mehdi Hosseini, Analyst: Understood. Very clear. Thank you.

Wendell P. Weeks, CEO and President: Okay.

Christopher Keenan, Investor Relations: Okay, thank you everyone for your participation. Before we wrap up, I'd like to let you know that we will be attending the Citi 2026 Global TMT Conference on September 9th. Additionally, we will be scheduling management visits to investor offices in select cities.

Finally, a webcast replay of today's call will be available on our website starting later this morning. Thank you again for your participation. Operator, this concludes the call. Please disconnect all lines.

Operator: Thank you. You may now disconnect.