The future of artificial intelligence cannot be separated from the physical systems that support it. Data infrastructure is the unseen backbone of this transformation, determining how quickly and securely information can move, and how reliably AI can function in the places it is most needed. The United States must urgently rethink how and where it builds AI infrastructure, or risk falling behind in this technological race.
“Artificial intelligence is transforming the way we as humans cognitively interact with the real world,” says William Michael Hadala, Vice President of Business Development and Strategy at American Lithium Energy. “If the US does not embrace this technology to the fullest, we’ll see adversarial nations capitalize off of it. AI makes humans almost superhuman, but without the right infrastructure, its power is wasted.”
At the center of his vision is GridEdge AI, a decentralized approach to computing that places processing power close to where data is generated. It is a strategy that challenges the reliance on hyperscale data centers, which Hadala views as unsustainable both in terms of energy and security. A civil engineering veteran turned national security technologist, Hadala has built his career on designing infrastructure that can withstand the demands of mission-critical applications.
Moving Beyond Hyperscale
The global rush to build hyperscale data centers risks repeating history. Hadala compares it to the early days of mainframes, when massive machines concentrated computing power in a few locations before eventually giving way to smaller, distributed personal computers and servers that reshaped accessibility and resilience. “Hyperscale data centers consume enormous amounts of power that many communities simply don’t have,” he says. “We need to move into a decentralized model where edge computing for AI lives where critical infrastructure operates: power utilities, water utilities, health care systems.” For Hadala, the alternative is not just inefficient but dangerous. He warns of a “digital Pearl Harbor” scenario in which attacks on centralized facilities could cut off access to AI capabilities nationwide. By distributing smaller, modular data centers across local substations, cell towers, and other nodes, resilience can be built directly into the network.
Energy as the Enabler
Hadala’s experience in energy storage gives him a unique perspective on what makes GridEdge AI possible. Before joining American Lithium Energy, he co-founded TMGcore, where he pioneered modular data centers cooled by dielectric fluids, a method 4,000 times more efficient than air cooling. After selling TMGcore to Modine in 2024, he turned his focus to the role of batteries as the foundation of edge computing. “It’s the energy that’s the enabler of high-capacity compute,” Hadala says. “Without advanced battery technologies at the edge, data is useless.”
American Lithium Energy, where Hadala now leads strategy, is commercializing defense-grade batteries that are safe, high-performing, and built within NDAA-compliant supply chains. These systems can be paired with microgrids powered by natural gas, hydrogen, solar, or even geothermal energy, creating self-sustaining compute nodes capable of supporting AI workloads.
Building Technological Sovereignty
Resilient infrastructure also requires resilient supply chains. Hadala points out that most of the world’s batteries and rare earth metals come from Asia, a dependency that leaves the US vulnerable. “Should we go into war tonight or tomorrow,” he cautions, “the US may have 40 days’ worth of batteries in inventory.” To reduce this risk, the Department of Defense is pushing domestic manufacturers to secure materials within 12 to 18 months. Hadala sees this as an essential step toward technological sovereignty. “The NDAA is critical. It ensures that future AI-driven systems across land, sea, air, and space will continue to thrive under US and allied leadership.”
Prioritizing Commercialization
Hadala is also deeply focused on how innovation moves from research to real-world deployment, wary of what he’s come to call the “valley of death” where early stage companies become reliant on government research funding but fail to win commercial adoption. “What works best is bringing in commercial partners early,” he says. “If you don’t have your commercial customers in sight, you’re going to struggle to commercialize long term. Success comes from innovating with existing technologies in novel ways, securing purchase agreements that last not just months but decades.”
Securing the Future of GridEdge AI
The future of AI will be determined not just by algorithms, but by the infrastructure that supports them. Looking ahead, Hadala envisions GridEdge AI converging with post-quantum cybersecurity, immersive compute, and autonomous defense platforms to build resilient cities and safeguard national security. He sees the combination of decentralized infrastructure, advanced batteries, and next-generation encryption as the path to making American systems both scalable and secure. “We’re living in a world where quantum computers could eventually decrypt all known encryption,” he explains. “By wrapping post-quantum security around GridEdge AI, you’re putting next-generation security around traditional systems, making networks nearly impossible to hack.”
Readers can follow William Michael Hadala Jr.’s work to secure better digital infrastructure in the US on LinkedIn or on hiswebsite.
