A proposal to build a sovereign artificial intelligence computing facility in Alaska could fundamentally change how the U.S. secures critical computing power for national defense and research, moving away from reliance on private technology companies. Port MacKenzie, an industrial district in Alaska's Mat-Su Borough, is being championed as the ideal location for a multi-gigawatt AI campus that would operate under government control rather than commercial interests. Why Is the U.S. Vulnerable to Private AI Infrastructure? The federal government currently depends on private cloud providers like Amazon Web Services, Microsoft Azure, and Google Cloud to run many of its most advanced artificial intelligence systems. This arrangement accelerated innovation and allowed agencies to adopt AI tools quickly, but it creates a strategic vulnerability. When critical computing infrastructure for national defense depends on commercial providers, the government's ability to guarantee long-term control, reliability, and independence becomes uncertain. Companies may change policies, shift priorities, or limit certain uses of their models in response to regulatory pressures in other markets. Even without direct conflicts, the incentives of private firms and national security institutions are fundamentally misaligned. In an era when artificial intelligence is as important to national security as nuclear deterrence or satellite systems, this dependency poses real risks. What Makes Port MacKenzie Uniquely Suited for This Project? Port MacKenzie offers a rare combination of physical and logistical advantages that few places on Earth possess. The site includes thousands of acres of industrial land expressly designed for large-scale energy infrastructure, deepwater port access, and proximity to the Alaska gas pipeline. These features matter enormously because modern AI training clusters require enormous amounts of electricity, often hundreds of megawatts or more, with future systems potentially requiring multi-gigawatt computing campuses. The Alaska gas line and a gigawatt-scale sovereign AI compute facility would be mutually supportive. The gas line would gain a reliable, large-scale anchor tenant capable of committing to long-term power contracts of sufficient size to secure financing for construction, while the compute facility would secure a reliable long-term power generation source. How Could This Project Benefit Alaska's Economy and Infrastructure? - Job Creation: A sovereign compute facility would create thousands of construction jobs, permanent technical positions, and a new ecosystem of research partnerships with universities and national laboratories. - Energy System Strengthening: The infrastructure required to power and connect the facility would strengthen Southcentral Alaska's energy system, lower energy prices and the cost of living, and revitalize the economy by attracting new businesses and industries. - Strategic Positioning: Alaska would be established as a secure, critical node in the nation's technological infrastructure, similar to how the Manhattan Project and early missile test ranges were deliberately placed in remote regions for national benefit. The concept of locating strategic infrastructure far from population centers is not new in American history. The Manhattan Project laboratories, early missile test ranges, and many satellite facilities were deliberately placed in remote regions where they could pilot, iterate, and scale in secure environments for the benefit of the nation. What Are the Energy Challenges Facing AI Data Centers Nationwide? The broader technology sector is grappling with unprecedented power demands. Data centers consumed approximately 4.4 percent of total U.S. electricity in 2023, with projections suggesting that consumption could rise to between 6.7 percent and 12 percent by 2028. This represents a tripling of data center load growth over the past decade, with expectations to double or triple over the next several years. Hyperscalers, the large technology companies that build and operate extensive cloud and data center infrastructure, are demanding large, steady blocks of power on accelerated timelines that do not match the current pace of transmission buildout or the lengthy interconnection queue. The U.S. power sector has not confronted load growth at this pace and scale for decades. Major technology companies are responding by pursuing multiple strategies. Alphabet has signed multi-gigawatt clean energy and power curtailment agreements with utilities such as DTE to support future data center needs, while also launching large-scale biochar and recycling projects in partnership with AMP Robotics to target long-term carbon reduction. These initiatives aim to manage rising energy demands from AI and cloud infrastructure while addressing emissions and regulatory expectations. What Role Could Natural Gas Play in Powering AI Data Centers? As hyperscalers seek clean, firm, and available-around-the-clock power, natural gas combined cycle units paired with carbon capture and storage technology are re-entering mainstream consideration. NGCC+CCS refers to a natural gas combined cycle generating unit paired with carbon capture and storage technology that captures the plant's carbon dioxide emissions for permanent storage. This technology serves as a "bridge" solution that can provide reliability needs with meaningful emissions reductions. Natural gas combined cycle generation is dispatchable today, meaning it can provide power on demand, while carbon capture and storage serves as the decarbonization add-on that attempts to close the emissions gap. However, NGCC+CCS projects must overcome high capital costs, supply-chain constraints, equipment procurement delays, and uncertainty regarding capture performance at scale. The shift in corporate clean-energy commitments reflects this reality. Google's goal to achieve 24/7 carbon-free energy by 2030, along with similar commitments from Microsoft and Amazon, has moved the conversation from renewable power procurement to clean firm power procurement, particularly in regions where new load is arriving faster than transmission and interconnection can keep up. How Could a Sovereign AI Campus Balance National Security and Innovation? The Alaska proposal suggests that the United States does not need to choose between national security and commercial innovation. Instead, the country could build infrastructure that allows both to thrive. A sovereign high-performance computing campus designed from the ground up to support national security, research, and technological leadership would provide the government with clear control and reliability while allowing private sector innovation to continue in parallel. At a moment when Washington is debating how to balance private innovation with national security needs, the Alaska proposal offers a concrete alternative. Rather than depending entirely on commercial cloud providers, the government could maintain its own secure computing infrastructure while continuing to leverage private sector advances in AI and accelerated computing. The convergence of massive power demands, national security concerns, and the need for clean energy solutions creates a unique opportunity. Port MacKenzie represents a location where these challenges could be addressed simultaneously, establishing Alaska as an indispensable node in America's technological infrastructure while securing computing power that serves national interests first.
