Five priorities shape AI strategies beyond the Washington and Beijing duopoly, challenging self-sufficiency myths

The international AI landscape reveals a wide gap between political rhetoric and technical reality, with data showing that the global race is not as open as it is portrayed, but is confined to a bilateral monopoly led by the United States and China. In 2025, American institutions produced 59 prominent models compared with 35 from Chinese institutions, while more than 90 % of leading models originated from commercial firms rather than academic centers. This concentration extends to hard infrastructure: Nvidia controls over 60 % of global computing capacity, the United States hosts 5,427 data centers, a ten-fold advantage over its nearest rivals, and a single Taiwanese company, TSMC, manufactures almost all advanced chips. In terms of financing, U.S. private investment reached $285.9 billion in 2025 versus roughly $6.8 billion in the European Union, while China dominates the volume of research publications, citations and generative AI patents.

This structural imbalance forces other countries to redefine the notion of “technical sovereignty.” Declarations by several governments, including Saudi Arabia, France and India, about launching national models and local computing centers represent understandable steps, but the problem begins when the concept of sovereignty is reduced to an attempt to reproduce the entire supply chain locally.Thus, a model classified as national remains fundamentally dependent on foreign chips, external storage clouds, and imported software.Consequently, the true measure of sovereignty is not measured by impossible self-sufficiency, but by the nature of dependence and its understanding, the possession of alternative options, and a state's ability to negotiate and impose conditions on the use of its data without falling into the grip of a single supplier.

International institutions cannot be relied upon to compensate for this structural deficit. Although the United Nations has created new bodies, such as the Independent International Scientific Panel on AI, which includes 40 experts, and launched the Global Dialogue on AI Governance in Geneva, these platforms remain venues for discussion and presentation of scientific evidence, not regulatory or financial entities capable of building public technological alternatives. It suffices to look at the expected capital spending of Google alone, ranging from $195 billion to $205 billion in 2026 for dedicated chips, data centers and networks, to realize the impossibility of private-sector-funded international organisations competing.

Given this reality, a realistic national strategy is defined by five practical priorities: first, a precise inventory of systems and sensitive data and a understanding of the cost of switching providers if services are disrupted; second, building technical and legal expertise within government procurement agencies so they act as “smart buyers” that deconstruct contracts and evaluate products independently; third, establishing red lines for systems that must not be outsourced, such as defense and national-identity systems, while allowing public computing for researchers and local companies; fourth, adopting a “technology-neutral” policy that preserves maneuvering space and prevents reliance on a closed cloud architecture that is difficult to change later; and finally, regional clustering and leveraging power assets such as land, energy and water in collective negotiations with cloud providers and data-center operators.

These shifts place decision-making rooms and tech institutions in the Gulf, Egypt and the Arab Levant before a direct test: building local data centers and models will not achieve their goals unless accompanied by training procurement and regulatory teams capable of reviewing licenses, data terms and preventing monopolies. As a technical or institutional leader in the region, you must recognize that the contractual decisions made today with infrastructure providers will shape the real alignment and determine the cost of digital independence for years to come.