Corning $GLW, AI fiber optic revenues doubled in a year, yet stock remains over 50% below its peak: What is the market really worried about? Corning $GLW announced its third-quarter earnings pre-market on Tuesday, surpassing expectations across nearly every metric. Core revenue reached $5.08 billion, up 19% year-over-year, exceeding the market expectation of $5.03 billion; core EPS was $0.91, up 36% year-over-year, surpassing the expected $0.88. Guidance for the fourth quarter is core revenue of $5.2–5.3 billion and core EPS of $0.96–1.00, with the midpoint above the market consensus of $0.93. Most notably, the optical communications segment posted quarterly sales of $2.38 billion, up 44% year-over-year; segment net profit reached $532 million, up 80% year-over-year; enterprise network business grew 84%, and sales of generative AI-related products more than doubled year-over-year. However, according to https://t.co/FK7CXkBPXL, the stock initially rose about 3% pre-market before reversing to decline approximately 3%, closing at $123.94—still roughly 54% below its 52-week high of $271.78. Let’s first examine what this earnings report has answered. Over recent periods, market concerns around AI hardware centered on two key questions: Will capital expenditures peak? And has optical communications been overvalued? Corning’s CEO, Wendell Weeks, responded that growth in generative AI products “has outpaced GPU shipment growth.” This means Corning’s expansion isn’t merely riding on the number of GPUs—it’s driven by increasing quantities of fiber and connectivity components per AI server and data center. As AI clusters grow larger, higher-density fiber connections are required between GPUs, cabinets, and data centers, meaning fiber consumption per unit of computing power is rising. Segment profit growth of 80% significantly outpaced revenue growth of 44%, indicating improved economies of scale and product mix—optical communications is now Corning’s highest-margin growth engine. Another easily overlooked insight is the certainty of orders. In Q3, Corning signed four long-term customer agreements worth billions of dollars each, including deals with Verizon $VZ and AT&T $T. This confirms that fiber demand isn’t limited to AI data centers—telecom operators are also upgrading their backbone networks and data center interconnects. Management noted on the earnings call that opportunities in data center interconnects within the telecom business are growing. The company’s “Springboard” initiative targets annualized revenue of $20 billion by end-2026, $30 billion by end-2028, and $40 billion by end-2030. Based on the Q4 guidance midpoint of $5.25 billion, annualized revenue is already nearing $21 billion—ahead of schedule for its first-phase goal. So why hasn’t the stock risen? First, Corning’s valuation is no longer cheap; the market had already priced in substantial AI growth at its peak. Even after a 50% decline, some valuation models still consider it above fair value. Second, growth is uneven: Display and glass innovation revenues declined 3% year-over-year; solar business capacity ramp-up lagged expectations, resulting in a quarterly loss of $7 million—both dragging down overall margin improvement. Third, the company has invested heavily in expanding optical communications capacity—Q3 capital expenditures reached $545 million—and the market needs to see this capacity absorbed at current price levels. Viewed within the broader AI optical communications supply chain, Corning represents the upstream fiber and connectivity segment, operating on the same demand chain as companies like Lumentum $LITE and Coherent $COHR, which produce optical modules and lasers—but with a different business model: Optical modules are more vulnerable to technological iteration and price competition, while fiber and connectors rely more on long-term contracts and scale. On Monday, Lumentum surged about 8%; today’s Corning earnings provide a more fundamentals-driven validation for the same chain. Next, watch three key metrics: Can optical communications maintain over 40% year-over-year growth in Q4? Will long-term agreements continue to expand? And when will the solar business turn profitable? When a company’s AI business doubles in a year but its stock remains at half its peak, do you think the market is pricing in cyclical risks—or offering long-term investors a window to reassess?