Silicon Statecraft: How U.S.-Gulf AI Deals Project Power

The great-power contest is not unfolding on battlefields or carrier decks, but inside data halls cooled by air conditioning, far from America’s shores. Rows of servers and racks of graphics processing units now carry as much strategic weight as military bases once did. Each deal for cloud access or advanced chips is a form of statecraft, binding partners into one camp’s technology ecosystem while locking out the other.

The United States is using AI infrastructure — data centers, cloud controls, and compute access — as a tool of power projection in the Arabian Gulf. By tying investment and capacity to governance safeguards, Washington can align regional partners with its security preferences, crowd out Chinese platforms, and set the rules for how AI is built and deployed. But the leverage is fragile. Without resilience and enforceable compliance, these arrangements risk becoming single points of failure or, worse, conduits for adversaries.

To make this new form of statecraft durable, U.S. policymakers should establish standard deal architectures with Gulf partners that combine hard technical safeguards, strict governance requirements, and built-in contingency plans. That means binding model weights to secure enclaves, tracking accelerators and workloads, embedding snapback clauses for violations, and pairing technical assurances with human rights standards. Done right, this approach can turn American-backed AI infrastructure into a lasting source of influence — quiet, scalable, and harder to dislodge than a forward operating base.

Data Centers Are the New Military Bases

In 2024, Microsoft took a $1.5 billion stake in Abu Dhabi’s G42. The deal came with an American board seat and explicit governance assurances. White House officials said the partnership was positive, since it pushed G42 to unwind ties with Huawei. This is AI statecraft in practice, capital plus compliance that aligns a partner’s tech stack with U.S. security preferences. As the United States reduces troop levels and closes or consolidates military sites across the Middle East, a network of U.S.-backed data centers and cloud controls is stepping in to do quiet work that bases once signaled.

Why the Gulf, and Why Now

The region offers cheap energy for compute, sovereign wealth funds willing to spend at scale, and governments that want AI capacity fast to diversify their economies. Saudi Arabia launched Humain under the Public Investment Fund in May 2025. Abu Dhabi’s MGX, backed by Mubadala and G42, is exploring plans to raise up to $50 billion in third-party capital. The real why is leverage. The Gulf can anchor an American technology stack in a strategic crossroads where partners are hedging between Washington and Beijing, which means the first mover sets the standards and the dependencies.

Capital Meets Compute, Then Policy

Gulf money is courting Silicon Valley labs, and U.S. platforms are building the pipes. The hinge was regulation. First, Executive Order 14110 directed the Commerce Department to advance AI safety and security measures, including proposals to collect information on certain foreign training runs and to safeguard model weights. Second, separate from Executive Order 14110, the Commerce Department published an infrastructure as a service proposal in January 2024 under earlier authorities (Executive Order 13984) that would require customer identification programs and reporting when foreign persons use U.S. cloud services to train large AI models. Together, these steps raised compliance burdens for deals, slowing Gulf partnerships even as guardrails improved. In 2025, the White House shifted toward quicker approvals for trusted partners while a replacement framework was still being developed. In May, the Department of Commerce rescinded the prior AI diffusion rule and issued interim guidance while a replacement rule is developed, signaling a simpler, ally-focused posture, but leaving specifics still in flux. The practical effect is that partners scale on American terms, and the rules travel with the hardware and the cloud.

What Executive Order 14110 Did, and Why It Mattered

Executive Order 14110 set AI safety and security priorities and directed the Department of Commerce to pursue measures that increased oversight of sensitive computing and model development, including information collection on certain foreign training runs, protection of model weights, mandatory reporting of red team results for dual-use foundation models, and notifications for large-scale computing clusters. These directions raised the compliance bar for foreign customers on U.S. clouds by adding disclosures, attestations, and review steps around high-risk training activity. In the Gulf, where speed, locality, and predictable capacity dominate, those added steps changed incentives. Saudi and Emirati data regimes already leaned toward in-country hosting, especially for government and regulated sectors, which amplified the appeal of local capacity. Huawei’s Riyadh region then offered immediate locality and a clear procurement path, giving buyers a ready option when U.S. pathways involved additional reviews. The result was not a documented wave of migrations but a procurement tilt: With Executive Order 14110’s added scrutiny and a viable in-region alternative, some customers treated Chinese platforms as the practical first choice for schedules that could not slip.

In practice, these dynamics increased delivery risk and schedule uncertainty on U.S. clouds, and procurement teams used that leverage in negotiations. An in-kingdom option with fewer U.S.-specific touchpoints and firmer commitments on timelines, service-level agreements, and data residency shifted price, terms, and technical defaults away from U.S. providers. Even when workloads stayed put, the perceived risk of delay reviews encouraged multi-cloud hedges, including consideration of non-U.S. vendors, to keep projects on track. In short, this executive order strengthened U.S. guardrails, but in the Gulf it also opened negotiating space that Huawei was positioned to occupy first.

What Changed Under Trump

On Jan. 23, 2025, President Donald Trump issued “Removing Barriers to American Leadership in Artificial Intelligence.” This revoked Executive Order 14110 and told agencies to suspend or rescind actions taken under it if they conflicted with the new policy. That shift deprioritized a centralized cloud reporting regime, favored streamlined licensing over preemptive restrictions, and assumed that competitiveness with allies increases security if China-facing lanes remain tight.

In May 2025, the Commerce Department said it would rescind and replace a pending global diffusion rule that would have divided countries into tiers and capped shipments of advanced AI chips. The stated aim was to move toward a simpler replacement framework, with enforcement and Chinese-focused tightening, while a full rule was still pending. Public reporting described the move as a loosening of planned curbs outside adversary jurisdictions. That meant more accelerators can reach friendly Gulf networks sooner, subject to safeguards, which accelerates their AI buildouts on an American stack.

At the same time, policy toward China did not relax. On Sept. 2, 2025, Washington revoked Taiwan Semiconductor Manufacturing Company Limited’s fast-track “Validated End User” status for shipments to its Nanjing fab, forcing case-by-case licenses even for mature nodes, with similar waivers for Samsung and SK Hynix due to expire. The subtext is clear: Ease the path for allies, tighten the valves on China, and steer partners toward a credible U.S. alternative to Huawei-centric ecosystems.

Policy Clarification

The Biden era: Bureau of Industry and Security export controls expanded in 2022 and 2023, and Executive Order 14110 added cloud reporting direction. The strict posture constrained adversaries, but it also risked ceding market share and influence to China where partners wanted capacity faster than U.S. processes allowed.

The Trump era so far: Executive Order 14110 revoked, and the administration is replacing the diffusion rule with a simpler, more permissive system for allies while keeping Chinese channels tight. The intent is to compete harder, undercut Chinese platforms, and use AI infrastructure to project U.S. power in the Middle East.

How The Leverage Works

When a partner accepts U.S. safeguards, American influence grows. Microsoft’s investment in G42 paired capital with a Huawei unwind. It also required governance changes, deeper audits, and Azure as the operating platform. The chip approval for the Microsoft-run Emirati site tied access to license conditions, including restricting facility access by personnel from embargoed countries or listed entities. In each case, model access, data location, and administrator privileges become policy levers that can be audited or revoked. These arrangements are not treaties. They are enforceable commercial and regulatory frameworks that travel with the infrastructure, and in doing so, shape market leverage through cloud governance and policy choices. Over time, those frameworks create dependencies that make U.S. standards the default and give Washington real veto power when security lines are crossed.

Benefits

Done right, the upside is straightforward. Partners get sovereign clouds built to U.S. security standards, and Washington gains platforms where audit, logging, and emergency cutoffs are possible. Arabic-first models improve services in health, energy, and finance, which reduces the appeal of Chinese alternatives because local needs are met on U.S. platforms. Scholarships, joint labs, and cloud credits draw researchers into U.S. tools and ecosystems, which turns training and habit into long-term alignment.

Risks

As Alex Rough argued in these pages, data centers are a battlefield, and doctrine has not kept pace. In the Gulf, these data centers are both targets and instruments of power. Leverage comes from control of chips, identity, and model weights, but it survives only if the nodes are resilient. Without that resilience, the same facilities that extend U.S. influence become single points of failure that invite cyber intrusion, AI-enabled sabotage, supply chain compromise, and cable cuts. A major outage would break partner trust and hand China the narrative advantage.

Resilience risks sit alongside governance risks that can hollow out leverage from within. Compliance can look strict on paper yet fail in practice. Weak custody of model weights or sloppy key management can move crown jewel models into uncontrolled spaces. Third-party integrators can reintroduce restricted vendors or exfiltrate know-how, and once advanced accelerators and weights mix in the wrong rack, control is hard to reassert. If partners use AI for mass surveillance or repression, pressure from Congress will push to cut access, and the result will be lost influence and fractured cooperation.

A Practical Policy Package

If Washington wants durable leverage that lasts without losing technical advantage, the deal architecture with Gulf partners should be standard and enforceable. Keep frontier model weights in hardware-bound enclaves with independent attestations, log every access, and make the logs reviewable by U.S. auditors. Tie chip exports to attestation, location tracking, and runtime telemetry so we know where accelerators sit, who runs them, and when workloads cross red lines. Build automatic snapbacks into contracts so compute quotas and sensitive features ratchet down if restricted vendors, personnel, or jurisdictions touch the stack, and restore only after fixes are verified. Attach human rights audits and clear acceptable use standards to financing and cloud credits and publish summaries so political support holds. Share the assurance burden by funding partner red teams, incident response, and compliance labs, and certify local teams to U.S. standards. Treat cloud infrastructure as contested terrain, as Alex Rough argues. In every deal, embed multi-cloud redundancy, automated failover to allied zones, and prearranged public-private response authorities so legal and operational seams are closed before a crisis.

The Gulf’s Strategic Role

Humain’s launch, MGX’s capital plans, and a steady flow of U.S.-Gulf tech announcements show where momentum sits. The Gulf combines sovereign capital, low-cost power, and political will. Strategically, it is the anchor point for an American-aligned AI corridor that links Europe, the Middle East, Africa, and South Asia, a place where standards and dependencies can be set for a generation.

Bottom Line

The contest over AI infrastructure in the Gulf is not just about market share or short-term influence. It is about who defines the standards of trust, control, and resilience for a technology that will underpin economies and security systems alike. Washington’s advantage lies in its ability to pair cutting-edge compute with enforceable governance, but that advantage will only hold if the United States treats data centers as contested terrain and invests accordingly. The test ahead is whether American policymakers can turn commercial deals into strategic architecture — one resilient enough to withstand crisis, competition, and the temptations of partners to hedge. If they succeed, silicon statecraft will give the United States a form of power projection suited to the century now taking shape.

Kody McKinley is a U.S. Army special operations veteran with over 12 years of service. He led teams on tactical deployments across the Middle East and later served with the Cyber National Mission Force as an information operations planner, conducting full-spectrum operations against advanced persistent threats and emerging technology threats. His analysis on Iranian proxies in Iraq has appeared in The Cipher Brief. He holds an M.S. in national security from Liberty University.

Image: Midjourney