Trump’s ‘Promoting Advanced Artificial Intelligence Innovation and Security’ Executive Order Ignores the Energy Security Challenge Behind AI Dominance

President Donald Trump’s new executive order, “Promoting Advanced Artificial Intelligence Innovation and Security,” is being framed as a necessary step to strengthen America’s cybersecurity posture and maintain leadership in the global AI race. The order focuses heavily on protecting critical infrastructure, hardening government systems, and creating mechanisms for voluntary cooperation between frontier AI developers and federal agencies.

Those are worthwhile objectives.

However, there is a glaring omission that could ultimately undermine both America’s AI ambitions and its national security objectives: energy.

The executive order repeatedly references cybersecurity, critical infrastructure, national competitiveness, and AI dominance. Yet nowhere does it meaningfully address how the United States plans to power the massive wave of AI infrastructure that will be required to achieve those goals.

For a document that seeks to strengthen America’s future, it largely ignores the foundation upon which that future will be built.

AI’s Real Bottleneck Is No Longer Compute

For years, discussions about AI leadership focused on chips, talent, and algorithms.

Today, the conversation has changed.

The limiting factor for many frontier AI projects is increasingly electricity.

Hyperscale data centers are consuming unprecedented amounts of power. The next generation of AI models will require even larger computing clusters, which in turn require enormous amounts of reliable energy. Utilities across the United States are already struggling to accommodate proposed AI facilities, and developers are increasingly competing for scarce grid capacity.

If America intends to lead the world in AI, energy security must become a core pillar of AI policy.

Instead, the executive order treats AI infrastructure largely as a cybersecurity challenge rather than an energy challenge.

Energy Security Is National Security

The administration is correct that advanced AI capabilities have national security implications.

But energy independence and grid resilience are also national security issues.

An AI ecosystem dependent on increasingly strained electrical grids, emergency natural gas expansion projects, and aging transmission infrastructure is not a resilient ecosystem.

The AI race is often described as a competition between the United States and China. If that framing is accepted, then the ability to deploy massive computing resources without destabilizing the domestic energy system becomes a strategic advantage.

Every new data center that requires additional fossil fuel generation introduces long-term vulnerabilities, whether those vulnerabilities come from fuel supply disruptions, transmission constraints, or rising electricity costs for consumers.

The executive order could have acknowledged this reality by establishing incentives or requirements for AI infrastructure developers to secure renewable and dispatchable energy capacity alongside new data center projects.

Instead, the issue is largely absent.

Climate Justice Advocates Highlight What the Order Leaves Out

Critics have already pointed to the environmental blind spots within the executive order.

Mar Zepeda Salazar, Policy Director at the Climate Justice Alliance, argued that the order focuses heavily on competitiveness and security while failing to address the local consequences of rapid infrastructure expansion.

According to Salazar:

“No mandatory environmental review. No energy or water use disclosures. No Tribal consultation. No cumulative impact analysis. No legal protections for communities.”

Salazar further warned that accelerated AI and data center development could increase electricity costs, place additional pressure on water resources, and encourage further fossil fuel development in communities that already bear disproportionate environmental burdens.

Whether one agrees with every aspect of that critique or not, it highlights a legitimate concern: the executive order assumes that more AI infrastructure is inherently beneficial without sufficiently addressing how that infrastructure will be powered or what safeguards should accompany its deployment.

The Technology Already Exists

The most frustrating aspect of this debate is that solutions are already emerging.

One example is Exowatt, a renewable energy company focused specifically on powering AI-scale infrastructure.

The company’s Exowatt P3 platform captures solar energy, stores it as heat in a thermal battery, and converts that energy back into electricity on demand. The system is designed to provide dispatchable power around the clock, addressing one of the most common criticisms of solar energy: intermittency.

Unlike conventional solar installations that depend heavily on battery systems or grid support, Exowatt’s approach uses thermal energy storage to deliver electricity when needed, making it particularly relevant for data centers that require continuous operation.

The company has attracted significant attention from investors and industry leaders, raising substantial funding while positioning itself as an energy solution for AI infrastructure. Its stated mission is to provide modular renewable power systems tailored specifically for energy-intensive applications such as data centers.

Whether Exowatt ultimately becomes a dominant player remains to be seen.

That is not the point.

The point is that companies are already building technologies designed to solve the AI energy problem.

The federal government should be encouraging these solutions as aggressively as it encourages AI model development itself.

Big Tech Has No Excuse

The world’s largest AI companies are among the most valuable corporations in history.

Companies such as Google, Microsoft, OpenAI, and Meta are spending tens or even hundreds of billions of dollars on AI infrastructure.

These companies possess the financial resources to accelerate deployment of renewable and dispatchable energy systems alongside new data center construction.

If AI truly represents the defining technology of the twenty-first century, then powering AI responsibly should be viewed as part of the cost of doing business, not an optional sustainability initiative.

The conversation should no longer be whether renewable-powered AI infrastructure is possible.

The conversation should be why it is not being required.

America’s AI Strategy Needs a Fourth Pillar

The White House executive order correctly recognizes that cybersecurity matters.

It correctly recognizes that frontier AI models introduce new risks.

It correctly recognizes that maintaining American leadership in AI is strategically important

But leadership requires more than faster model releases and stronger cyber defenses.

It requires a plan to power the future.

A truly comprehensive AI strategy would rest on four pillars: innovation, security, infrastructure, and energy resilience.

Without the fourth pillar, the first three become increasingly difficult to sustain.

The United States can absolutely lead the world in artificial intelligence. But if policymakers are serious about long-term AI dominance, future executive orders should not merely ask how America can build more data centers.

They should ask how America intends to power them for the next several decades.

The answer to that question may ultimately prove just as important as the AI models those facilities are built to run.