Strategic Autonomy in the Age of Artificial Intelligence

Strategic Autonomy in the Age of Artificial Intelligence

ResearchPosted on rand.org Jul 15, 2026Published in: Geopolitics of AI: Power, Conflict, and the Future of Global Order, Chapter 1, pages 13-30 (2026)

The post-Cold War era was shaped by the assumption that economic integration and openness would diminish strategic rivalry and bind states into a durable framework of shared prosperity and restraint.

That assumption has unraveled, alongside the demise of the post-Cold War international order.

An increasingly common assumption today is that deep interdependence in an increasingly contested, competitive, and chaotic world is a vulnerability, not an asset. Strategic autonomy, in turn, has become the organizing principle of our geopolitical moment. For most countries, this strategy seeks to reduce exposure to strategic dependencies, enhance freedom of action, and protect and strengthen domestic sources of national competitiveness. For some, it is also animated by an aspiration to become an independent pole in a more multipolar world.

This essay argues that the rapid advance of AI will frustrate ambitions for strategic autonomy. It develops the concept of the geopolitical AI stack—a map of the layered capabilities required to leverage AI as a means of enhancing national power and influence—to challenge several prevailing strategic assumptions. “Middle powers” that anticipate greater relative weight in international politics will find that the gap with the great powers may grow rather than shrink. Leading powers that anticipate translating AI advantage into geopolitical dominance will find that sustained dependencies on external resources and other players, among other factors like adoption challenges and the societal trials of the AI transition, pose formidable obstacles. The defining challenge ahead is not to escape dependency but to manage it wisely: building strong national capacity across the stack while shaping the cooperative architectures that an era of transformational AI—full of civilizational risks and opportunities—will demand.

This publication is part of the RAND external publication series. Many RAND studies are published in peer-reviewed scholarly journals, as chapters in commercial books, or as documents published by other organizations.

RAND is a nonprofit institution that helps improve policy and decisionmaking through research and analysis. RAND's publications do not necessarily reflect the opinions of its research clients and sponsors.