Middle-power countries are rapidly adopting artificial intelligence (AI) in military operations but remain dependent on AI infrastructure controlled by superpowers like the United States and China, exposing them to significant strategic risks.
What happened
The escalating conflict between the U.S. and Iran since 2026 marks the first large-scale deployment of advanced AI in military operations, with notable incidents like Iranian drone attacks on Amazon Web Services data centers serving multiple clients. This highlights the intertwined nature of civilian commercial AI infrastructure and military action. Meanwhile, American companies such as SpaceX and Planet Labs have made operational decisions—such as restricting satellite imagery or internet service—that critically affect wartime capabilities in other countries, underscoring the power consolidated in private firms tied to U.S. interests.
Middle powers, defined as countries with significant military ambitions but lacking a full domestic AI stack, face a dependency trap. They rely on foreign hardware, cloud platforms, AI models, and software primarily from U.S. and Chinese suppliers. This dependency restricts their sovereignty because these foreign components embed design priorities and operational constraints aligned with their home countries’ strategic goals.
An illustrative example is India’s acquisition of French Rafale jets without access to source code critical for radar and electronic warfare functions, paralleling the “black box” problem in AI systems. As a result, middle powers risk having their military AI capabilities degraded or disabled remotely through “kill switches” or covert throttling by providers or their home states, especially under conflict conditions where timing and precision are crucial.
Efforts by countries like India, South Korea, and Indonesia to build sovereign AI are constrained by the global concentration of AI research talent, computing power, and investment—over 70% of researchers and 90% of global computing power are controlled by the U.S. and China. Even sovereign AI models rely on foreign chips and cloud infrastructure, resulting in managed dependency rather than true technological independence.
Why it matters
The reliance of middle powers on foreign AI infrastructure in defense threatens their strategic autonomy and ability to independently conduct military operations. This dependency could allow supplier countries to influence or inhibit middle powers’ military effectiveness without overt conflict, through infrastructure control or legal pressure on AI suppliers. It also complicates geopolitical alignments and defense cooperation since middle powers may be forced to conform to the strategic priorities of their technology providers.
This dynamic intensifies geopolitical tension at a time when military AI capabilities are becoming central to modern warfare. The risk of “soft kill switches,” such as deliberate latency increases or capability downgrades, introduces new vulnerabilities that could decide outcomes in conflict zones without traditional combat engagements. It highlights the urgent need for international governance frameworks addressing the operational use of military AI and protections for dependent states.
Background
The integration of AI into military systems relies on complex technology stacks comprising hardware, software, data, and cloud services, predominantly built and controlled by U.S. and Chinese firms. These suppliers respond primarily to their home governments’ policies and strategic interests. Middle powers lack the industrial base and resources to develop and maintain independent AI ecosystems at parity with the superpowers.
Past experiences, such as India’s restrictions on modifying French military technology due to sealed source codes, demonstrate similar dependency challenges pre-dating AI. The stakes have risen with the growing operational role of AI, which is rapidly becoming intrinsic to targeting, surveillance, and command decisions in military conflicts.
Attempts at “sovereign AI,” including India’s IndiaAI Mission and South Korea’s $75 billion investment, mostly address governance and accountability at the software level but cannot fully eliminate reliance on foreign AI chips and cloud infrastructure. Participating in alliances such as the Pax Silica Declaration ties middle powers closer to U.S.-led AI supply chains, exposing them to ongoing dependency framed as strategic cooperation.
Sources
This article is based on reporting and publicly available information from the following source:
Read more World News stories on Goka World News.