Proposal: Transforming AT&T’s Legacy Central-Office Infrastructure into a National Neutral-Host Data-Center Platform
Executive Summary
The proposed acquisition of AT&T for an estimated total transaction value of approximately $340 billion would be based not only on the value of AT&T’s telecommunications operations, but also on the strategic value of its nationwide portfolio of central offices, real estate, electrical infrastructure, and fiber-optic connectivity.
The central concept is to separate AT&T’s telecommunications operations from ownership of the underlying real estate and infrastructure, creating a separate real-estate and data-center infrastructure business modeled in part on the neutral-host infrastructure approach used by companies such as American Tower and Crown Castle.
Rather than requiring AT&T to continue owning and maintaining thousands of aging central-office properties, qualifying properties could be transferred or sold into a separate real-estate investment and infrastructure structure. The infrastructure company would develop modern data centers on underutilized land surrounding existing central offices, particularly parking lots that are no longer needed as AT&T’s workforce and legacy operations decline.
AT&T would then lease back the space and infrastructure it requires, while the same facilities could be made available to other telecommunications companies, cloud providers, artificial-intelligence companies, and enterprise customers.
The result would be a nationwide, neutral-host data-center platform built around existing electrical connections, fiber-optic networks, and strategically located telecommunications facilities.
- The Central-Office Opportunity
AT&T’s central offices were built over many decades to support a telecommunications network that was fundamentally different from today’s network.
Historically, these facilities housed large, energy-intensive switching systems and supported extensive copper telephone networks. Many buildings were designed to accommodate equipment and electrical loads that were significantly greater than what modern fiber-based telecommunications equipment requires.
As AT&T continues to retire legacy copper infrastructure and older switching technology, the telecommunications function of these locations will not disappear. The fiber-optic network and modern network equipment will remain essential.
However, the amount of physical space and electrical capacity required by AT&T’s remaining telecommunications operations may be substantially lower than the capacity historically required by the legacy network.
This creates a potentially valuable mismatch: large existing electrical and telecommunications infrastructure versus a much smaller modern telecom equipment footprint.
The opportunity is to redirect the resulting excess capacity and underutilized real estate toward new uses.
- Build New Data Centers on Existing Property
The proposal is not to convert the aging central-office buildings themselves into data centers.
Many of these buildings are decades old and may contain asbestos and other legacy construction materials. They may also be inefficiently designed for modern high-density computing.
Instead, the initial strategy would be to use the underutilized land surrounding the central offices, particularly parking lots that are no longer needed because the buildings require fewer employees.
Modern, purpose-built data centers would be constructed on these sites.
The new facilities would be designed specifically for modern computing requirements, including:
- AI computing
- High-performance computing
- Cloud services
- Enterprise computing
- Edge computing
- Telecommunications infrastructure
- Data storage
The existing central-office buildings would continue operating during the transition, maintaining AT&T’s fiber and telecommunications infrastructure.
As legacy switching and copper equipment are retired, AT&T’s remaining equipment could eventually be consolidated into a much smaller, modern telecommunications footprint.
At that point, the obsolete central-office buildings could potentially be demolished and the land incorporated into the data-center campus.
- Existing Electrical Infrastructure as a Strategic Asset
One of the most important potential advantages of this strategy is the existing electrical infrastructure.
Central offices were historically designed around substantial electrical requirements. They were equipped with utility connections, transformers, electrical distribution systems, backup power, batteries, and generators to support mission-critical telecommunications operations.
As legacy switching equipment and copper infrastructure are retired, the electrical demand associated with the old network should decline substantially.
The objective would be to identify locations where significant electrical capacity can be made available for new data-center development.
Each central-office property would therefore be evaluated based on:
- Existing utility service capacity
- Transformer capacity
- Substation proximity
- Available electrical capacity
- Existing backup generation
- Battery systems
- Utility expansion potential
- Local electricity costs
- Available land
- Fiber-optic connectivity
- Long-haul network access
- Local permitting and zoning
- Proximity to major population centers
- Potential cooling resources
The most attractive locations could become major regional data centers, while smaller locations could support smaller edge-computing facilities.
- The Neutral-Host Data-Center Model
The central element of the proposal is to separate ownership of the infrastructure from the telecommunications companies and other businesses that use it.
This is similar in concept to the model used by companies such as American Tower and Crown Castle.
Wireless carriers such as AT&T, Verizon, and T-Mobile do not necessarily need to own every tower on which their equipment is installed. Instead, an independent infrastructure company owns the tower and leases space to multiple competing carriers.
The same principle could be applied to data-center infrastructure.
Instead of AT&T owning every central-office property and every future data-center building, a separate infrastructure company or real-estate investment structure would own the properties and facilities.
AT&T would become a tenant.
Other telecommunications companies could also become tenants.
Cloud providers, artificial-intelligence companies, enterprise customers, and other businesses could become tenants as well.
The infrastructure would be neutral.
For example, a future central-office/data-center campus could have:
- AT&T leasing space for telecommunications equipment
- Verizon leasing data-center capacity
- T-Mobile leasing data-center capacity
- Cloud providers leasing computing or colocation space
- AI companies leasing high-density computing capacity
- Enterprise customers leasing infrastructure
- Other telecommunications and technology companies leasing space
The infrastructure owner would provide the physical environment, including:
- Land
- Buildings
- Electrical infrastructure
- Cooling infrastructure
- Physical security
- Data-center space
- Network connectivity
Customers would pay for the infrastructure they use.
This would transform the former AT&T central-office network from a proprietary real-estate portfolio into a neutral-host national infrastructure platform.
- AT&T Becomes a Tenant of Its Former Real Estate
Under this structure, AT&T would no longer need to own and maintain the majority of the real estate associated with its central-office network.
Instead, AT&T would lease back the space it needs.
For example, a future site could consist of a modern data center in which the majority of the facility is available for commercial tenants while a smaller portion is reserved for AT&T telecommunications infrastructure.
AT&T would pay rent under a long-term agreement.
The telecommunications network would continue operating normally, but AT&T would no longer bear the full cost of owning and maintaining the entire property.
This would allow AT&T to focus capital on its core businesses:
- Wireless
- Fiber
- Broadband
- Enterprise telecommunications
- Network services
Meanwhile, the real-estate and infrastructure company would focus on:
- Data-center construction
- Electrical infrastructure
- Property management
- Leasing
- Colocation
- AI infrastructure
- Cloud infrastructure
- Potential Corporate Structure
A possible structure could involve three distinct components.
AT&T Operating Company
Responsible for:
- Wireless services
- Fiber network
- Broadband
- Enterprise customers
- Telecommunications operations
- Network equipment
Real-Estate and Infrastructure Company
Responsible for:
- Central-office properties
- Data-center campuses
- Electrical infrastructure
- Buildings
- Physical infrastructure
- Leasing
Data-Center Operations
Responsible for:
- Colocation
- AI computing
- Cloud infrastructure
- High-performance computing
- Enterprise data-center services
Depending on tax and regulatory requirements, the real-estate component could potentially be structured as a REIT, potentially with taxable REIT subsidiaries or other affiliated entities where necessary.
The exact structure would require detailed tax, securities, telecommunications, and regulatory analysis.
- Separating Management Responsibilities and Incentives
A critical element of the proposed structure would be the separation of the telecommunications operating company from the real-estate and digital-infrastructure business.
The skills required to operate a mature telecommunications company are fundamentally different from those required to develop a rapidly expanding national data-center and artificial-intelligence infrastructure platform.
The telecommunications operating company would continue to focus on:
- Wireless networks
- Fiber networks
- Broadband
- Enterprise telecommunications
- Customer service
- Network reliability
- Telecommunications technology
The infrastructure company, by contrast, would focus on:
- Data-center development
- Electrical infrastructure
- Artificial-intelligence infrastructure
- High-performance computing
- Real-estate development
- Construction
- Capital allocation
- Leasing
- Rapid technology deployment
This separation would allow each business to be managed by executives with the appropriate expertise and incentives.
The objective would be to avoid imposing a traditional telecommunications management structure on a business that is fundamentally different from traditional telecommunications.
A nationwide data-center platform operating in the rapidly evolving AI economy requires a management culture that is entrepreneurial, technologically sophisticated, aggressive in capital deployment, and capable of making decisions quickly.
The infrastructure company could therefore recruit leadership from the technology, infrastructure, and entrepreneurial sectors rather than assuming that the existing management structure of a traditional telecommunications company is best suited to lead a new national AI infrastructure platform.
The goal would be to bring in leadership with a demonstrated ability to move rapidly, challenge conventional assumptions, control costs, and execute large-scale infrastructure projects.
The underlying principle is simple:
Let telecommunications professionals run the telecommunications network.
Let infrastructure and technology professionals build and operate the data-center platform.
The separation would also create a more transparent measurement of performance. AT&T’s telecommunications management could be evaluated on the performance of the telecom business, while the infrastructure company’s leadership could be evaluated on data-center construction, occupancy, return on invested capital, and long-term infrastructure value.
This structure could also provide the flexibility to recruit an unusually strong technology leader for the infrastructure company—potentially someone with the entrepreneurial and execution-focused qualities demonstrated by leaders such as Elon Musk, or another leader capable of moving dramatically faster than a traditional telecommunications organization.
The objective would not necessarily be to replicate any particular individual or company. Rather, it would be to bring the same type of entrepreneurial and execution-focused leadership that has allowed certain technology companies to move dramatically faster than traditional incumbents.
The underlying principle is that the opportunity created by AT&T’s infrastructure may be significantly larger than what a traditional telecommunications management structure would naturally recognize or pursue.
By separating the businesses, the new ownership could prevent the infrastructure opportunity from becoming subordinate to the priorities of the legacy telecommunications organization.
Instead, the data-center and infrastructure platform would have its own leadership, capital allocation, and strategic objectives.
In effect, the transaction would create two fundamentally different businesses from one existing company:
AT&T — the telecommunications operating company.
A national neutral-host digital infrastructure company — focused on power, real estate, data centers, AI, and technology infrastructure.
Each could then be operated by management specifically selected for the business it is responsible for building.
- Why the Strategy Could Be Valuable
The primary advantage is that the strategy could reduce the time and capital required to develop new data-center infrastructure.
A new data center typically requires:
- Land acquisition
- Utility negotiations
- Electrical interconnection
- Substation development
- Fiber connectivity
- Permitting
- Zoning
- Construction
The AT&T portfolio potentially provides many of these components already.
The properties are already associated with:
- Telecommunications infrastructure
- Utility connections
- Fiber-optic networks
- Network routes
- Established commercial properties
- Existing security
- Existing backup power infrastructure
The strategy therefore attempts to convert legacy telecommunications infrastructure into modern digital infrastructure.
The old copper network created the need for thousands of strategically located facilities.
The next generation of technology—AI, cloud computing, and high-performance computing—could potentially reuse the geographic and infrastructure advantages created by that earlier network.
- A Nationwide Distributed Data-Center Network
The ultimate objective would be to create a national network of data centers rather than relying exclusively on a small number of massive hyperscale campuses.
Different properties could serve different purposes.
Large Sites
Potentially supporting:
- 50–200+ MW data centers
- AI training
- High-performance computing
- Hyperscale cloud operations
Medium Sites
Potentially supporting:
- 10–50 MW facilities
- Regional cloud computing
- Enterprise computing
- AI inference
Smaller Sites
Potentially supporting:
- 1–10 MW facilities
- Edge computing
- Low-latency applications
- Telecommunications
- Regional AI inference
This distributed model could provide computing capacity closer to customers and population centers while utilizing AT&T’s existing nationwide fiber network.
- The Long-Term Vision
The ultimate vision is to transform AT&T’s legacy central-office footprint into a nationwide neutral-host digital infrastructure network.
The transformation would be:
Legacy copper network
→ Modern fiber network
→ Reduced telecom equipment footprint
→ Underutilized electrical capacity and real estate
→ New purpose-built data centers
→ AT&T becomes a tenant
→ Verizon, T-Mobile, cloud providers, AI companies, and enterprises become tenants
→ Nationwide neutral-host digital infrastructure platform
The historical telecommunications network would therefore provide the foundation for the next generation of digital infrastructure.
Instead of allowing thousands of central-office properties to become stranded assets as copper and legacy switching disappear, the properties could be redeveloped into productive infrastructure supporting the growth of artificial intelligence, cloud computing, and the broader digital economy.
- The Core Investment Thesis
The fundamental investment thesis is not simply to acquire AT&T as a telecommunications company.
It is to acquire a combination of:
Telecommunications operations + nationwide fiber network + strategic real estate + electrical infrastructure + data-center development potential.
The telecommunications business would continue operating, but the legacy real-estate model would be fundamentally changed.
The old central-office buildings would gradually be replaced by modern facilities. The land currently occupied by unnecessary parking lots would become new data-center campuses. AT&T would lease the space it needs, while competitors and third-party technology companies would be able to lease the remaining capacity.
The model would be similar in principle to the neutral-host tower industry pioneered by companies such as American Tower and Crown Castle, but applied to the rapidly expanding demand for data centers and AI infrastructure.
The result could be a nationwide infrastructure company that owns the land, power, buildings, and physical infrastructure, while multiple competing telecommunications and technology companies pay to use that infrastructure.
The key question for further analysis is therefore not simply the value of AT&T’s existing telecommunications business.
The key question is:
How much unused electrical capacity, strategically located real estate, and fiber connectivity exists across AT&T’s central-office network—and what would that infrastructure be worth if redeveloped into a nationwide neutral-host data-center platform?
That is the question that should drive the next stage of the analysis.