Verizon Communications has signed a dark fiber agreement worth more than $1 billion with Alphabet’s Google, marking one of its largest artificial intelligence infrastructure contracts to date and positioning the U.S. telecom giant to capitalize on the multitrillion-dollar investment wave reshaping the technology sector.
The agreement, announced Friday during Verizon’s second-quarter earnings call, will see the telecommunications company provide dark fiber connections linking Google’s expanding network of AI data centers across the United States.
The deal reveals a rapidly emerging investment theme in the AI economy: while semiconductor companies such as Nvidia have dominated headlines, the enormous computing demands of artificial intelligence are also creating significant opportunities for telecommunications firms, fiber network operators and other providers of critical digital infrastructure.
Register for Tekedia Mini-MBA edition 20 (June 8 – Sept 5, 2026).
Register for Tekedia AI in Business Masterclass.
Join Tekedia Capital Syndicate and co-invest in great global startups.
Verizon Chief Executive Dan Schulman described the agreement as a pivotal moment for the company, saying it reflects a broader strategy to become a key connectivity partner for hyperscale cloud providers as they accelerate AI infrastructure spending.
“The build out of AI infrastructure across the United States is one of the largest capital cycles of our lifetime, and Verizon is uniquely positioned to participate in it,” Schulman said.
Investors welcomed the announcement. Verizon shares rose more than 3% in early trading, while Alphabet’s Class A shares gained about 1%.
Dark fiber refers to unused optical fiber infrastructure that companies lease to customers, who then install and operate their own networking equipment. Unlike traditional managed telecommunications services, dark fiber provides customers with dedicated, high-capacity and low-latency connections that can be scaled as computing demands increase.
The technology has become increasingly valuable as artificial intelligence models require enormous volumes of data to move rapidly between geographically dispersed data centers, cloud regions and computing clusters.
Training and deploying advanced AI models involves connecting thousands, and in some cases tens of thousands, of graphics processing units (GPUs) across multiple facilities. Those workloads require ultra-high-bandwidth, low-latency fiber networks capable of transmitting vast amounts of data continuously, making fiber infrastructure an increasingly strategic asset in the AI ecosystem.
Schulman indicated that the Google agreement is only the beginning of Verizon’s broader AI infrastructure strategy. He said the company expects to announce additional agreements before the end of the year that together could generate several billion dollars in revenue over the coming years.
Those contracts, he said, are expected to be long-term in nature, providing Verizon with predictable cash flows while serving some of the world’s largest technology companies.
Schulman described Friday’s announcement as “consequential,” saying it signals the direction in which Verizon’s future revenue growth is headed.
The move is seen as an effort to monetize assets that were originally built to support traditional telecommunications services but have become increasingly valuable in the age of artificial intelligence. Verizon operates one of the largest fiber networks in the United States, including extensive long-haul routes connecting major metropolitan areas and metro fiber systems serving urban data centers and enterprise customers.
According to Schulman, that infrastructure is now ideally positioned to meet the connectivity requirements of hyperscale AI developers. The company’s network, originally designed for an earlier generation of internet traffic, has become well suited for linking AI data centers, high-performance computing clusters and cloud regions as technology companies rapidly expand their computing capacity.
The agreement also underpins the unprecedented scale of AI-related capital expenditure currently underway. Major technology companies including Google, Microsoft, Amazon, Meta and OpenAI-backed infrastructure projects are collectively committing hundreds of billions of dollars annually to build new AI data centers, acquire advanced semiconductors and expand supporting infrastructure.
While much investor attention has focused on chipmakers and cloud providers, analysts increasingly see networking infrastructure as one of the most important bottlenecks in AI deployment. Without sufficient fiber capacity, the computing power housed inside AI data centers cannot be efficiently connected, limiting the performance of distributed AI workloads.
That dynamic is creating new opportunities for telecommunications companies that own extensive fiber assets.
For Verizon, the Google partnership also represents a diversification of its revenue base. Like many traditional telecom operators, Verizon has faced slowing growth in its core wireless business as the U.S. mobile market has matured. Expanding into AI infrastructure allows the company to leverage existing network investments to tap into one of the fastest-growing areas of enterprise technology spending.
The contracts are also attractive from a financial perspective because they typically involve long durations, high switching costs and investment-grade counterparties, characteristics that can provide stable recurring revenue over many years.
But the agreement also supports Google’s accelerating AI expansion plan. The company continues to invest aggressively in new data centers and computing infrastructure to support its Gemini AI models, cloud services and enterprise AI offerings. As AI workloads become larger and more geographically distributed, reliable high-capacity fiber connections have become as important as access to advanced semiconductors.
The first phase of the AI boom largely rewarded semiconductor manufacturers and cloud computing providers. Increasingly, however, the next phase is benefiting companies that provide the underlying physical infrastructure, including utilities, power equipment manufacturers, cooling system suppliers and fiber network operators.
As AI models become more computationally intensive, demand for high-speed connectivity between data centers is expected to grow alongside demand for electricity and advanced chips.
If Verizon secures the additional multibillion-dollar contracts that Schulman indicated are in the pipeline, the company could establish itself as one of the leading connectivity providers for the next generation of AI infrastructure, opening a significant new avenue for long-term growth as hyperscalers continue investing at record levels.