Nokia's two moves are not aimed at short-term orders, but at the dominance of optical networks in the AI era. In mid June, Nokia released two heavyweight messages in less than 48 hours: On June 16th, it was announced that the advanced packaging capacity of photon chips at the Allentown factory in Pennsylvania would be expanded tenfold; On June 18th, it was announced that it will collaborate with T3 Broadband and Aureon to establish an initial 100 Tb/s and scalable 400 Tb/s ultra high capacity optical transmission network between North Dakota and Chicago. Looking at these two things together, it's not just about capacity expansion and project implementation, but rather Nokia's systematic layout in the AI infrastructure field. The core logic is to simultaneously grasp supply chain security and real commercial verification. 1. Allentown expands production: the "localization" trump card for American AI optical networks The Allentown factory is currently one of the few facilities in the United States capable of packaging photonic chips into optical modules, with the United States accounting for less than 2% of the world's Advanced Semiconductor Packaging (ATP). Nokia has invested $30 million this time (including CHIPS Act support), with the goal of increasing production capacity by 10 times, and the new capacity will be available by the end of Q3. This is not simply 'expanding production', but establishing local manufacturing capabilities in key tracks. The demand for optical modules in AI data centers is exploding, but the supply chain is highly dependent on Asia. Nokia's move can not only respond to the US government's demand for localization of key technologies, but also reduce the geopolitical and tariff risks when supplying hyperscalers to the US in the future. More importantly, it is highly compatible with Nokia's ongoing "AI native network" strategy - optical modules have lower power consumption and higher integration, directly serving the ultimate requirements of energy efficiency and bandwidth for the next generation of AI clusters. Executive Justin Hotard directly defined this as a 'direct investment in the AI super cycle'. 2. Aureon Project: Transforming Technical Validation into a replicable Business Template The ultra high capacity optical transmission network from North Dakota to Chicago has an initial capacity of 100 Tb/s and an expandable capacity of 400 Tb/s, using the Nokia 1830 GX platform and 1.2T ICE7 coherent optics. The significance of this route is that personal search validation is likely to be the "first" 100 Tb/s+project in North America, and it is also a commercial landing driven by real AI data center development projects. The current North American DCI market is in the stage of transitioning from "urban high fiber core parallelism" to "long-distance single fiber ultra-high capacity". According to Dell'Oro data, the optical transmission market is expected to grow by 10% year-on-year to around $16 billion by 2025, with DCI related revenue increasing by approximately 40% year-on-year. In 2026, the overall market will continue to grow rapidly, and the growth rate of DCI will be significantly higher than the average. Nokia's use of the latest generation coherent optics and C/L band technology in AI load scenarios is like giving the market a shot in the arm: our solution works in real high load environments. This has great reference value for scale across DCI orders similar to the Midwest Corridor project and even Hyperscaler in the future. 3. Comparison with Ciena: Nokia is filling the most critical link In the DCI segment, Ciena has long been a leader, and the WaveLogic series has a higher penetration rate in hyperscalers, with a stronger focus on pure optics. But with Nokia's acquisition of Infinera to fill the gap in related engines and vertical integration capabilities, the gap is narrowing. This local production capacity of Allentown combined with a real North American DCI landing case is equivalent to simultaneously addressing Ciena's relatively weak links in supply chain security and commercial verification. Ciena is more like a 'tech pure player', while Nokia is a combination of 'technology+manufacturing+ecology'. In the long run, this combination may be more competitive in the US market, especially as hyperscalers and operators increasingly focus on supply chain diversification and localization. 4. The real impact on Nokia Optical Networks Optical networking has become the fastest-growing sector in Nokia Network Infrastructure. Q1 Optical Networks increased by 20% year-on-year, with significant contributions from AI&Cloud orders. The company has raised its 2026 growth target for optical+IP networks to 18-20%. Allentown's expansion aims to address the issue of 'whether it can be supplied', while the Aureon project addresses the question of 'whether others are willing to buy'. The combination of the two is a substantial benefit for subsequent order conversion. Especially in the high gross profit and fast-growing market of DCI in North America, if Nokia can continue to win similar reference cases, the weight of Optical business in the company's overall profitability will be further enhanced. Of course, competition remains fierce. Ciena will not easily give up its share, and the trend of hyperscalers making their own pluggable modules continues. But Nokia has now obtained clear differentiation labels in at least two dimensions: "US domestic+real AI scenario verification". In the longer term, this is the layout for 2027-2028: Nokia's two moves are essentially betting on the dominance of optical networks in the AI era. Allentown is a long-term layout of production capacity and supply chain, while Aureon is a short-term validation of technology and market. The combination of the two sends a clear signal: Nokia not only wants to sell devices, but also wants to become a systematic player in the "key components+reliable delivery" of AI infrastructure. For the industry, such actions will further accelerate the evolution of North American DCI (Data Center Interconnection) towards higher capacity and lower power consumption, and will also force other players to follow suit in local manufacturing and real-world verification. For Nokia itself, if this combination of "production capacity+verification" can be continuously replicated in the future, the Optical Networks business is likely to become one of its most certain growth curves in the AI cycle. This is not the story of one or two projects, but a milestone in Nokia's transition from a "participant" to an "important player" in the optical field.