Quick Buzz Feed
Nvidia has emerged as a dominant player in sovereign AI, which accounts for roughly 14% of the company’s revenue, according to a Stanford HAI study.
Big tech companies are tapping into unmet needs around operational control with their global cloud and AI sovereignty initiatives, while deepening countries’ and companies’ reliance on their products, a study from the Stanford Institute for Human-Centered Artificial Intelligence found. The study reviewed offerings from the most dominant companies with extensive global reach — Microsoft, Google, AWS, Nvidia and OpenAI. Adopting Nvidia’s sovereignty offerings, for example, could drive vendor lock-in as switching to rivals such as Intel or AMD would be costly, the study found. OpenAI’s sovereign AI might increase local access to frontier models, but doesn’t allow further control over underlying systems. Finally, while Google, Microsoft and AWS have increased their sovereign cloud offerings, they can be costly and complex to operate, according to the analysis. “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,” the study said.
Rising regulatory risks are pushing enterprise leaders to weigh their options when it comes to AI infrastructure, including the benefits of sovereignty offerings. But big tech’s product lineup may not be all they promise, according to Stanford HAI. Three-fourths of business leaders fear the geopolitical risks of storing their data in global cloud environments, according to Kyndryl’s 2025 Cloud Readiness Report. Meanwhile, 65% of business leaders are changing their cloud strategies due to digital sovereignty requirements largely promulgated in countries outside the U.S. Governments in Europe and other countries globally are not only implementing sovereignty requirements, but investing for their own AI tools and services. The European Union unveiled the European Technological Sovereignty Package last month in a bid to curb reliance on U.S. tech companies. Though European policymakers in particular have demonstrated a willingness to invest in homegrown alternatives, Microsoft, AWS and Google have responded by bolstering their own sovereignty offerings to companies in European countries, offering what Stanford HAI identified as the “largest and most global set of solutions.” Global sovereign cloud spend is expected to increase 35.6% this year, according to a Gartner report. While sovereign AI products offered by big tech companies do address concerns about control over AI supply chains, they often reconfigure rather than eliminate dependencies and could result in greater entrenchment — a dilemma especially highlighted by cross-stack sovereignty products like Nvidia’s AI factories, according to Stanford HAI. Nvidia describes its AI factories as locally owned and operated AI clouds for training and inference that lean on public-private partnerships to scale infrastructure and meet the innovation needs of countries and companies. Additionally, Nvidia said its AI Nations initiative has assisted countries in building out their AI ecosystems. France’s Ministry of Economy and Finance, for example, is using AI agents built on the Nvidia AI platform — through infrastructure controlled in the country — to automate complex workflows and process millions of documents, according to Nvidia. The vendor has begun to position itself as a “one-stop-shop” infrastructure company, offering sovereignty at the foundational layer through its AI factory program, the Stanford HAI study said. Sovereign AI accounts for 14%, roughly $30 billion, of Nvidia’s total revenue, according to the report. “They may promise greater control and integration across different layers of the AI tech stack, but they can also tighten long-term vendor lock-in, reducing interoperability and hardening reliance,” the study said.