An Open Letter to the Industry : A Call for Collaboration on AI Data Center Infrastructure Standards

Driven by the AI revolution, the data center construction boom is now a major economic force, contributing $355 to $727 billion in annual GDP between 2017-2023 and matching consumer spending’s contribution to GDP growth over the first half of 2025. These investments are injecting hundreds of billions of dollars directly into the economy for construction, specialized labor, and the procurement of hardware like GPUs, advanced networking, and liquid cooling systems. This scale of investment necessitates not just an increase in capacity, but a fundamental, strategic evolution in data center design.

To enable rapid data center deployments, a fundamental rethinking of how physical infrastructure is built is required. A primary goal for hyperscalers is to enable late-binding decisions, such as deploying GPUs or other accelerators within a given facility. However, this is difficult when data center partners receive competing design inputs from various hyperscalers, enterprise users, and technology providers. This lack of standardization leads to inefficiencies, slows down deployment, and ultimately hinders innovation. To keep pace with the exponential growth in demand, we need to move towards common infrastructure standards within the data center industry that encompass an interoperable model for data center infrastructure.

To meet the challenges and opportunities of this new era, we, the undersigned, believe it is time for a new level of collaboration across the data center industry.

A Framework for Collaboration

This effort is not about creating a single, one-size-fits-all solution, but rather about identifying key areas where we can agree on common interfaces and specifications while still allowing for differentiation and innovation. This will require a concerted effort from all stakeholders, including hyperscalers, colocation providers, and key technology partners.

Momentum for industry collaboration in these important areas is building, with significant progress made in the first half of 2025. A series of workshops (1, 2, 3) that were held as part of the OCP Open Systems for AI Strategic Initiative successfully united a diverse group of stakeholders—from hyperscalers and chip manufacturers to power, cooling, and colocation providers—to initiate critical conversations. Our collaborative work on specifications like Mt Diablo and Embodied Carbon Disclosure showcase this progress.

To accelerate our progress, we propose establishing collaborative workstreams to define the foundational elements of an efficient and adaptable data center, beginning with four key areas:

  • Power: With AI racks soon to surpass 500 kW, power distribution is likely to require high-voltage DC systems that disaggregate components out of the IT rack. Immediate collaboration should focus on standardizing the interfaces for these systems. Over time, the intent is to leverage Solid State Transformers (SSTs) to fully integrate the power distribution system at the facility level, eliminating the need for separate sidecar power racks. Efforts should include defining acceptable thresholds and standards between utilities and data centers, progressing existing efforts to support grid stability, and identifying new potential solutions. The long-term vision is a resilient, end-to-end power ecosystem.
  • Cooling: This same increase in power density makes liquid cooling essential. Immediate collaboration on Coolant Distribution Units (CDUs) should focus on standardizing their screening and commissioning processes. The long-term goal is to align on a common CDU design, starting with a core set of shared mechanical interfaces and telemetry standards.
  • Mechanical: The sheer size and immense weight of highly interconnected AI systems has surpassed traditional spatial design of general purpose compute and storage. This creates significant spatial challenges, making many current data halls incompatible. Immediate collaboration must focus on standardizing key physical parameters— including aisle widths, rack dimensions, and weight-bearing capacity—to ensure nextgeneration hardware can be deployed without costly retrofits. The long-term vision is a common physical structure that enables the seamless, often with robotic assisted operations, placement of any compliant rack, achieving true fungibility at the facility level.
  • Telemetry: As data center infrastructure becomes more disaggregated and complex, the lack of a common, secure language for monitoring and control creates significant operational risks. Immediate collaboration must focus on defining a standard protocol for telemetry and control that incorporates security from the ground up. The long-term vision is to enable autonomous control systems to coordinate data center power consumption and load fluctuations. This will enable a unified control plane that can orchestrate all physical systems to optimize performance and enable self-healing capabilities.

Alignment on these key interfaces creates a win-win scenario. Data center providers can support a wider range of customers with common, reusable infrastructure, protecting their large capital investments. At the same time, customers benefit because standard interfaces—rather than rigid, prescriptive designs—encourage faster innovation across the ecosystem.

These foundational workstreams are just the beginning. Future efforts will need to address grid interactions, battery energy storage systems (BESS), and beyond. By collaborating on these initial goals, we can start a virtuous cycle, building the trust and momentum needed to move faster, protect our collective investments, and build the infrastructure required for the future. The Path Forward

This is an ambitious undertaking, but we believe it is essential for the future of our industry. We have a choice: continue with fragmentation or come together to build a more open and innovative future.

We plan to use the Open Compute Project’s flexible collaboration framework to make rapid progress, and we invite you to join us. By creating a critical mass behind this effort, we can reduce new capacity cycle times, speed up the adoption of new technologies, and build a more efficient data center ecosystem for everyone.

Let’s work together to define the flexible and efficient data center infrastructure that will serve our industry for the next decade.

Sincerely,

Sign the Letter today to Join the Companies Above!