Verizon has found a new way into the artificial intelligence boom, and it does not involve building another chatbot or launching a large language model. The US telecommunications company has secured an agreement worth more than $1 billion to provide dark fiber connectivity for Google’s expanding data center network. Verizon CEO Dan Schulman disclosed the deal during the company’s second-quarter earnings call on July 24, 2026.
It is a major contract on its own. Verizon, however, sees it as the beginning of something much larger. The company expects to announce several additional infrastructure agreements before the end of 2026. Those deals could generate several billion dollars in revenue over the coming years as technology companies spend heavily on the physical systems needed to train and run AI models.
Google Is Paying for the Network Behind Its AI Expansion
Google’s AI ambitions require more than processors, servers and enormous data center buildings.
Those facilities must move huge amounts of information between computing clusters, cloud regions and storage systems. Slow or congested network connections can limit the performance of even the most advanced AI hardware. That is where Verizon enters the picture.
Under the long-term agreement, Verizon will provide dark fiber connecting Google data centers. Dark fiber refers to installed optical fiber that does not yet carry active signals. The customer can equip and operate the fiber according to its own capacity, security and performance requirements.
For Google, that means greater control over the network connecting parts of its AI infrastructure. For Verizon, it turns existing fiber assets into a high-value service for one of the world’s largest technology companies.
The AI Infrastructure Race Is Moving Beyond Chips
AI infrastructure conversations usually begin with graphics processors.
NVIDIA chips remain central to the industry. Google also continues to expand its own Tensor Processing Unit ecosystem. Yet chips cannot operate in isolation. AI companies need electricity, cooling systems, land, data centers and reliable network connections capable of carrying increasingly large workloads. The Verizon-Google agreement highlights that less glamorous layer of the AI economy.
Fiber networks do not attract the same attention as a new model launch. They are still essential. Training clusters may contain thousands of processors working together, while inference services must respond quickly to users spread across different locations. Every prompt, image, video and model response creates network traffic somewhere. As AI systems grow, moving that information efficiently becomes a business opportunity of its own.
Verizon Wants to Become a Major AI Connectivity Provider
Verizon has been preparing for this shift through its AI Connect strategy.
The company introduced the offering to provide high-capacity connectivity, edge computing resources and infrastructure services for cloud providers, hyperscalers and enterprises deploying AI workloads. Its network stretches from 5G infrastructure and fiber routes to edge locations that can support computing closer to end users. Verizon originally built many of these assets for telecommunications services. AI data centers have given them another purpose.
Schulman said the company’s network now fits the requirements of businesses connecting data centers, computing clusters and geographic regions. He described the Google contract as the first of several long-duration opportunities Verizon is pursuing. That wording matters. Verizon is not treating the deal as a one-time fiber sale. It wants recurring infrastructure revenue from companies building AI systems at enormous scale.
More Billion-Dollar Deals Could Arrive Before Year-End
Verizon expects to announce additional AI infrastructure agreements by the end of 2026. The company has not identified the potential customers or disclosed how many contracts remain under negotiation. Schulman said the combined opportunities could produce multiple billions of dollars in revenue over the next several years.
These agreements could give Verizon a steadier revenue stream than parts of its traditional consumer business. Large data center projects typically involve extensive planning, dedicated infrastructure and long-term commitments. Once a network connection becomes part of a customer’s core architecture, replacing it is neither quick nor simple.
That creates the kind of durable relationship telecom providers want. It also puts Verizon in competition with other fiber operators, cloud connectivity companies and infrastructure specialists chasing the same demand.
The Deal Arrives During a Verizon Turnaround
The Google agreement was announced alongside stronger-than-expected second-quarter results from Verizon. The company added 184,000 postpaid phone subscribers during the quarter. Mobility and broadband service revenue increased, although total revenue slipped as customers purchased fewer devices.
Investors responded positively to the results and the company’s move into AI infrastructure connectivity. Verizon also raised its service revenue expectations while maintaining its broader financial outlook. The timing is useful. Verizon has spent years trying to convince investors that its network can generate growth beyond conventional mobile subscriptions. A billion-dollar agreement with Google gives that argument more weight.
Google Keeps Spending on the Physical Side of AI
Google has been expanding nearly every part of its AI infrastructure. The company is investing in custom chips, cloud capacity, data centers and the networking systems that connect them. It has also pursued large infrastructure partnerships as demand for AI training and inference continues to climb.
The dark fiber agreement with Verizon fits inside that broader strategy. Google can develop powerful models and specialized processors, but those resources still need to operate as part of a fast, coordinated network. AI performance increasingly depends on the entire system rather than one piece of hardware. That system includes the fiber beneath the ground.
What the Verizon Google AI Infrastructure Deal Means
The biggest AI contracts are no longer limited to software companies and semiconductor manufacturers. Telecommunications groups, construction companies, energy suppliers, cooling providers and data center operators are all becoming part of the AI supply chain. Verizon’s agreement with Google offers a clear example of how far that spending is spreading.
For Verizon, the deal opens a potentially lucrative business around unused or underutilized network assets. For Google, it secures dedicated connectivity for infrastructure that will handle increasingly demanding AI workloads. And for the wider industry, it delivers another reminder: the AI race will not be won through models alone. Someone still has to build the physical network that keeps them running.
Sources
- PYMNTS: Verizon Lands $1 Billion Google Deal as AI Infrastructure Demand Soars
- Reuters via Yahoo Finance: Verizon Signs $1 Billion Dark Fiber Deal for Google Data Centers
- Verizon: AI Strategy to Power Next-Generation AI Demand
- Light Reading: Verizon Lands $1 Billion AI Data Center Connectivity Deal With Google
