How will fiber and equipment vendors meet the increased demand for fiber optics in 2026 due to AI data center buildouts?

Fiber optic vendors are employing a mix of manufacturing expansion, technological innovation in high-density and next-generation fibers, and strategic supply chain alignment to meet the anticipated surge in demand from AI and data centers in 2026. The demand is so high that at least one major fiber manufacturer,  whose name was not explicitly disclosed in news reports, has already sold all its fiber inventory through 2026.  Major fiber optic vendors by category are:

  • Fiber & Cable Manufacturing: Corning, Prysmian Group, Sumitomo Electric, Fujikura, CommScope, Sterlite Technologies (STL), Yangtze Optical Fibre & Cable (YOFC)
  • Optical Transport/Networking: Nokia, Ciena (gaining share), Cisco, Fujitsu, Huawei, Infinera (now part of Nokia)
  • Optical Components/Transceivers: Coherent Corporation, Lumentum, Broadcom, Innolight, Accelink

Major focus areas of selected vendors:

  • Corning: Leading in fiber cable quality and innovation
  • Nokia & Ciena: Strong in optical transport and network solutions, gaining market share
  • Cisco & Huawei: Significant players in optical transceivers, catching up to leaders
  • CommScope, Clearfield, STL: Preparing for huge demand surges

Selected Quotes:

John McGirr, SVP and general manager for Corning Optical Fiber & Cable, said, “The surge in hyperscale and AI network loads has significantly increased our expectations for fiber demand. Enterprise sales grew 58% year-over-year in Q3 2025, driven by continued strong adoption of Corning’s Gen AI products, largely due to AI network growth demands.  The 72-GPU nodes, such as (Nvidia’s) Blackwell, require 16 times more fiber than traditional cloud switch racks. We see no signs of AI network growth slowing down especially as operators scale up (increase computational power by adding more resources within the existing backend AI network node) and scale out (increase the number of nodes to accommodate increasing demand) their networks.”

Rahul Puri, CEO of the Optical Networking Business at STL, said, “AI-focused data centers require significantly more fiber —  about 36x more fiber than traditional CPU-based racks — to handle the massive data volumes and high-speed connectivity required by GPU clusters.” Puri predicts that cumulative hyperscale data capacity will increase by three times in the next few years alone. “The U.S. will need to add 213.3 million more fiber miles by 2029, more than doubling its current amount from 159.6 million fiber miles to 372.9 million miles. Our roadmap is shaped directly with the world’s leading cloud, AI and data center operators,”” Puri added.

CommScope’s VP of Technology John Chamberlain and VP of Hyperscale Cloud Erik Gronvall noted that the company has expanded its fiber manufacturing capacity in recent years to meet increased demand. “We are also innovating to reduce the amount of time it takes to deploy AI clusters,” said Chamberlain and Grovall.

………………………………………………………………………………………………………………………………………………………

Fiber Vendor Strategies:

Capacity Increase: Vendors like Corning and CommScope are investing in increasing their production capacity for fiber optic cables and the necessary preforms (raw material for fiber). This includes expanding existing facilities to help alleviate the current supply chain tightness and long lead times.

Technological Innovation in Fiber Design: To support the extreme bandwidth and low-latency needs of AI, vendors are focusing on advanced fiber technologies.

    • Higher Fiber Counts: Companies are launching cables with extremely high fiber counts (e.g., 1,728+ strands) and higher density options to pack more capacity into existing infrastructure.
    • Next-Generation Fibers: Research is ongoing in areas like hollow-core fiber (which uses air or a vacuum to transmit light faster and with less loss) and multicore fiber (multiple cores in one strand to increase capacity). These technologies, while not yet mainstream for 2026, are part of the long-term strategy.
    • Bend-Insensitive Fiber: Innovations in bend-insensitive and ultra-high fiber count cables are improving durability and easing deployment in complex data center environments.

Pre-connectorized and Modular Solutions: To counter a persistent skilled labor shortage and speed up deployment, vendors are pushing factory-terminated, plug-and-play fiber systems and modular platforms (like Siemon’s LightStack). These solutions require less on-site expertise and reduce installation time.

Strategic Partnerships and Supply Chain Alignment: Vendors are forming strategic collaborations with hyperscalers and network operators (like the agreement between Corning and Lumen) to align manufacturing platforms with future demand and ensure supply. They are also working to optimize supply chains and, in some cases, regionalize manufacturing to reduce lead times.

Structured Cabling and Photonics: There is a renewed focus on structured cabling architectures, as recommended by some AI platform providers, to ensure predictable, low-latency performance and simpler long-term management. The industry is also exploring integrated photonics to address the power and thermal challenges of future systems.

Focus on AI-Specific Demands: Vendors recognize that AI data centers require up to five times more connectivity than traditional hyperscaler topologies and network architectures. Their strategies are specifically tailored to high-volume, intra-bay, inter-bay, and middle-mile fiber connections to link distributed data center clusters into a single, unified AI computing environment.

…………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………

Ciena and Nokia:

Fiber optic network equipment vendors like Ciena and Nokia are preparing for increased demand in 2026 by significantly ramping up production of high-speed optical components (like 800G and 1.6T), investing heavily in R&D and manufacturing capacity, and developing new network architectures specifically optimized for AI workloads.
1. Ciena is focusing on scaling its production and developing solutions for both short-reach internal data center connections and long-haul links between geographically distributed data centers. 
  • Ramping up Production: Ciena is accelerating the production of 800G ZR+ optical pluggables, with plans to ship a large volume in 2026 to major cloud providers who are currently testing the technology.
  • New Architectures: The company is developing new interconnect solutions under the “Scaleacross” architecture designed to support growing AI workloads by significantly increasing capacity and density within the data center.
  • Increased Forecasts: Driven by record orders from hyperscalers, Ciena has raised its revenue guidance for fiscal 2026 to a range of $5.7 billion to $6.1 billion, a significant increase that analysts tie directly to AI-driven demand.
  • Strategic Positioning: Ciena emphasizes that the network will be the primary limiter of AI performance by 2026, positioning its high-speed fiber solutions as critical for moving massive amounts of data between compute nodes efficiently. 
2. Nokia is undergoing a major strategic overhaul to position itself as a leader in AI networking, focusing on a holistic approach that covers the entire network infrastructure. 
  • Major U.S. Investment: Nokia announced a $4 billion investment in U.S. R&D and manufacturing capabilities for “AI-ready” network technologies, including optical and data center networking, to ensure robust domestic supply.
  • Strategic Reorganization: Effective at the start of 2026, Nokia will reorganize into two primary segments, one of which is “Network Infrastructure” (including optical networks), which it sees as the center of the “AI supercycle.”
  • Industry Collaboration: Nokia has deepened its commitment to the Open Compute Project (OCP) at the Platinum level, aiming to collaborate on open, interoperable AI networking innovations that optimize space, cost, and power efficiency with standards-driven technology.
  • Advocacy for Network Modernization: Nokia’s research highlights that current networks are insufficient for future AI growth, advocating for substantial investment and cross-industry collaboration to modernize digital infrastructure to handle the uplink-heavy, distributed data flows generated by AI. 

…………………………………………………………………………………………………………………………………………………………….

References:

https://www.fierce-network.com/broadband/major-fiber-vendors-strategize-huge-demand-ai-2026

https://www.fierce-network.com/premium/research/1410126?pk=FN-Research-Commscope-111925-listing

NTT to launch 25 Gps FTTH service in Tokyo starting March 2026

AI wireless and fiber optic network technologies; IMT 2030 “native AI” concept

AI infrastructure investments drive demand for Ciena’s products including 800G coherent optics

AT&T sets 1.6 Tbps long distance speed record on its white box based fiber optic network

China Telecom with ZTE demo single-wavelength 1.2T bps hollow-core fiber transmission system over 100T bps

Lumen and Ciena Transmit 1.2 Tbps Wavelength Service Across 3,050 Kilometers

Co-Packaged Optics to play an important role in data center switches

Coherent Optics: Synergistic for telecom, Data Center Interconnect (DCI) and inter-satellite Networks

Hyper Scale Mega Data Centers: Time is NOW for Fiber Optics to the Compute Server

Microsoft acquires Lumenisity – hollow core fiber high speed/low latency leader

China Mobile to deploy 400G QPSK by the end of 2023

Leave a Reply

Your email address will not be published.

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

*