A new report by Stanford University’s Institute for Human-Centered Artificial Intelligence (HAI) observes that despite aggressive push by many countries for AI sovereignty, dependence on major US based tech companies remains strong as ever.
The report titled “The Commercial Landscape of AI Sovereignty Offerings” as July 2026 brief observes that companies such as Nvidia, Microsoft, Google, Amazon Web Services and OpenAI are offering AI sovereignty solutions to governments around the world, that include local data centres, secure cloud services, AI infrastructure and customized AI models.
The report states that while these services by US based tech companies give countries more control over their data and AI operations, their indigenous sovereign AI push gets in the backseat, as they get dependent on the tech and infrastructure outsourced.
The report explains the term “sovereignty whitewashing” and how US based tech/AI companies offering tools to other countries creates a dependency cycle.
“While the tools these companies sell under the banner of sovereignty may genuinely improve purchaser countries’ control over certain aspects of AI development and deployment..they also ensure these countries will remain structurally dependent on them for the long term,” observes the report.
The report highlighted that the idea of AI sovereignty gained attention globally after NVIDIA CEO Jensen Huang made a statement that “every region and every country needs to build their sovereign AI” at an AI conference in Paris in November 2023. He also made a similar remark at Dubai’s World Government Summit and later officially launched AI sovereignty initiatives in February 2024.
The study further mentioned that since then Nvidia has been promoting the term “AI sovereignty” and the idea that countries should prioritise national ownership of AI development, with the company promising to provide the tools and infrastructure to achieve their goal.
According to the report, reliable AI infrastructure is essential for countries seeking greater control over AI due to which many of these countries are racing to secure computing resources through massive capital investments, despite the high costs and environmental challenges involved.
“AI sovereignty should be understood as a state’s capacity to act deliberately and make independent decisions concerning the development, deployment, and governance of AI systems that fall along a spectrum of interdependent arrangements. When considering purchasing corporate AI sovereignty solutions and forming partnerships, decision-makers should prioritize expanding strategic choice without losing access to frontier capacity,” the report mentioned.
However, the report warned that although countries are investing heavily in AI infrastructure to strengthen their AI capabilities, many remain dependent on a single US chipmaker.
“US led export controls have left some countries on unstable footing or altogether unable to access the advanced hardware their sovereign AI ambitions require,” the study argued.
Also Read: Macron Pushes AI Alliance with India as U.S. Bans Claude Fable for Non-Americans
It cited an example of Kazakhstan which faced significant procurement delays for its first supercomputer because of US export controls on Nvidia chips.
The report also mentions the case of Sarvam AI, an indigenous platform of India advertised as “full-stack sovereign AI platform.”, commenting that in reality, the company focus their offerings on the model and application layers of the tech stack.
The report suggested that the “core policy challenge of AI sovereignty is not eliminating dependence but calibrating interdependence and sovereignty should be viewed as the capacity to shape and negotiate dependencies, not avoid them entirely.”
Stanford HAI has also given a disclaimer for the report stating it has received financial support from several companies referenced in this brief, including Nvidia, Google, Microsoft, and AWS.
