Corning Surges More Than 100% This Year: AI Optical Interconnect Revaluation for the 175-Year-Old Materials Giant

In 2026, NVIDIA remains the byword for AI chips in the capital markets, but a more covert investment narrative is taking shape—AI data centers’ physical-layer infrastructure is undergoing a paradigm shift from copper to optical. One of the biggest beneficiaries of this shift is a specialty glass and materials science company with a 175-year history: Corning Incorporated (Corning, NYSE：GLW).

As of June 17, 2026, Corning’s stock closed at $177.42. Although it pulled back from the 52-week high of $211.51 set in May, the year-to-date gain remains substantial. Multiple data sources show that GLW’s total return for the year ranges from 92% to 120%. Even more noteworthy is how it compares with peers: over the same period, the S&P 500 rose only about 6%, while semiconductor leaders such as NVIDIA and Broadcom gained between 14% and 51%. Corning not only outperformed the market benchmark, but also beat most AI chip stocks.

The 52-week trading range is $48.85 to $211.51—measured from the low, the maximum increase exceeds 330%. A company once known around the world for mobile phone glass is now completing a fundamental reappraisal at the base layer of AI infrastructure.

The “Nervous System” of AI Data Centers: Fiber Moves from Auxiliary Component to Core Bottleneck

To understand the logic behind Corning’s stock revaluation, it’s necessary to return to a basic physical question. AI training clusters consist of thousands of GPUs working in concert. On NVIDIA’s latest GPU platform, network communication speeds have reached 800Gbps and are moving toward 1.6Tbps. At these speeds, traditional copper cabling faces two insurmountable obstacles: signal attenuation and thermal management. In high-speed transmission, copper cables’ signal quality degrades sharply, and the heat generated can no longer be effectively dissipated in high-density racks.

Data center architects are forced into a fundamental change: moving fiber optics from “interconnecting data centers” to “directly connecting GPUs and switches inside server racks.” This is no longer an incremental upgrade, but a comprehensive replacement of the data center’s “central nervous system.”

According to CRU’s statistics, global demand for data center fiber optics is expected to reach 91.6 million fiber kilometers in 2026, up 32% year over year. By 2030, this figure will further grow to 128 million fiber kilometers, with fiber demand for AI applications exceeding 80 million fiber kilometers. More critically, there are structural constraints on the supply side: the expansion cycle for optical-fiber preform production lasts 18 to 24 months, and the process certification barriers are extremely high. Estimates indicate that in 2026 the global supply-demand gap for optical fiber and optical cables will be about 6%.

A $6 Billion Meta Order and a “Scramble” Among Three Giants to Secure Corning

On January 27, 2026, Corning and Meta Platforms announced that they had signed a multi-year agreement worth up to $6 billion to accelerate the construction of the most advanced data centers in the United States. Under the agreement, Corning will supply Meta with the latest generation of fiber optics, cables, and connectivity solutions. To support this demand, Corning will expand its manufacturing capacity in North Carolina—including a large-scale capacity expansion at its fiber cable manufacturing plant in Hickory—while Meta will serve as the anchor customer for the plant. Corning Chairman and CEO Wendell Weeks said that this investment will support a 15% to 20% increase in employment levels in North Carolina and sustain a highly skilled workforce of more than 5,000 people.

Meta is not the only hyperscale customer reaching a major procurement agreement with Corning. During the Q1 earnings call in April 2026, Corning disclosed that two additional long-term supply agreements with a scale and duration similar to the Meta deal are also underway. In May 2026, NVIDIA announced a long-term strategic partnership with Corning: Corning committed to expanding its U.S. optical interconnect manufacturing capacity by tenfold and expanding its U.S. optical fiber capacity by more than 50%. This expansion includes building three advanced manufacturing plants in North Carolina and Texas, which are expected to create more than 3,000 jobs.

In June 2026, Amazon also joined the trend, signing multi-billion-dollar fiber supply agreements with Corning to support its rapidly expanding U.S. data center infrastructure. The agreement is expected to create approximately 1,000 jobs at Corning’s manufacturing base in North Carolina.

Within just half a year, three of the four largest AI infrastructure buyers worldwide—Meta, NVIDIA, and Amazon—have signed long-term, large-value contracts with Corning. This concentration of customers sends a clear industry signal: fiber optics are no longer just the “pipelines” of data centers, but the “nervous system” of AI compute clusters.

Optical Communications Business: Profit Soars 93%, Contributing More Than Half of Core Net Profit

In the first quarter of 2026, Corning delivered results that exceeded market expectations. Core total revenue was $4.35 billion, up 18% year over year, exceeding analysts’ estimate of $4.29 billion. Core earnings per share were $0.70, up 30%, reaching the high end of the company’s guidance. Core operating margin expanded by 220 basis points to 20.2%.

Among them, the optical communications segment is the absolute growth engine. In Q1, the segment’s revenue was $1.8 billion, up 36%, accounting for 44.6% of total company revenue. Even more notable is profitability: net profit for the optical communications segment reached $387 million, up 93% year over year. This means the segment’s net profit margin rose from roughly 14% in the prior-year period to over 21%—a double effect of both scale benefits and pricing power. The segment currently accounts for more than half of the company’s $612 million core net profit.

In its earnings report, Corning CEO Wendell Weeks said, “Our strong first-quarter performance continues the powerful momentum of the Springboard plan,” and pointed out that AI is driving “the largest scale infrastructure buildout of our era,” adding that Corning’s expanded optical communications capacity will directly meet the connectivity needs required by ultra-large-scale data center deployments of NVIDIA accelerated computing.

Looking ahead to the second quarter, management expects core sales to grow by approximately 14% year over year to about $4.6 billion, and core EPS to increase by about 25% to the $0.73 to $0.77 range. The 2028 sales outlook has been raised by 25% to nearly $30 billion, and provides a framework for reaching $40 billion in revenue by 2030. Management has also proposed a grand plan to double sales to $40 billion by 2030.

GLW vs. NVDA: Who Outperformed Whom in 2026?

If Corning is compared with AI-chip leader NVIDIA, a counterintuitive conclusion emerges. So far in 2026, NVIDIA’s stock climbed to $216 at one point in late April, then quickly fell back. Meanwhile, Corning achieved gains of 92% to 120% over the same period. Multiple media outlets have described Corning as a “super semiconductor stock that crushed NVIDIA in 2026.”

Of course, this comparison is not intended to deny NVIDIA’s core position in AI compute power. NVIDIA GPUs remain the foundation for AI training and inference, but Corning’s rise reveals an important fact: beneficiaries of the AI value chain extend far beyond chip manufacturers. From optical modules to optical fiber and cables, from connectors to data center cabling, the entire physical-layer infrastructure is undergoing a value reappraisal. As AI clusters expand from a tens-of-thousands-of-cards scale to a hundreds-of-thousands scale, and networks evolve from “auxiliary components” to “core bottlenecks,” the companies that solve this bottleneck are capturing an attention-worthy share of the AI compute upside.

Gate Platform: Trade GLW U.S. Stocks Directly with USDT

For investors who focus on Corning as an AI infrastructure investment target, the choice of trading channel is equally important. On June 1, 2026, Gate officially launched real stock trading services, becoming one of the first exchanges in the industry to directly connect the U.S. stock market within a crypto platform.

As of June 2026, Gate TradFi has launched more than 11,500 real stocks and ETFs, fully covering five major exchanges including NYSE, Nasdaq, NYSE Arca, NYSE American, and BATS. Users only need to use the USDT liquidity in their Gate account to buy real stocks listed on the NYSE and Nasdaq with one click.

The core advantages of Gate’s real stock trading are reflected in three dimensions. First, fractional share trading has an extremely low threshold: you can start with as little as 0.01 share, and begin investing in U.S. stocks with $1. Second, settlement is done directly in USDT, completely eliminating the cumbersome process of “selling crypto → withdrawing fiat → cross-border remittance → depositing funds at a broker.” Third, full SIPC protection: all stock trades are executed by compliant broker-dealers in the United States that hold the relevant Broker-Dealer license and clearing eligibility, backed by real assets independently custodied under the DTC system.

This means users can complete allocation to U.S. stocks such as Corning (GLW) on the Gate platform without leaving the crypto ecosystem. For investors who are bullish on the long-term trend of AI infrastructure, this channel provided by Gate significantly lowers the barriers and friction costs of trading across markets.

Conclusion

From providing glass for Edison’s light bulbs, to supplying Gorilla Glass for Apple iPhones, and now providing the optical “nervous system” for AI data centers—over its 175-year history, Corning has completed industry transformations time and again. The 2026 story is particularly distinctive: AI doesn’t just need compute chips; it also needs the physical network that connects those chips. And within this network, Corning is making a value leap—from a “materials supplier” to a “provider of core AI infrastructure components.”

Meta’s $6 billion order, NVIDIA’s strategic partnership, and Amazon’s multi-billion-dollar agreements—three of the world’s largest AI infrastructure buyers signing deals with Corning in succession within six months—are not a coincidence. They are a direct reflection of the industry logic as AI compute cluster expansion accelerates: “the network as a bottleneck” is an inevitable reality.