The company said the growing adoption of generative and agentic AI has increased the urgency for countries to establish their own AI capabilities. Beyond supporting large language models, Nvidia said these systems can be used for software development, drug discovery, fraud prevention, robotics and speech technologies, including models designed to preserve indigenous languages.
Central to Nvidia’s strategy is the concept of “AI factories,” which it describes as next-generation data centers built to process large-scale AI workloads. According to the company, these facilities form the computing foundation for training and running AI models, with governments adopting different ownership and operating models, including partnerships with state-owned telecommunications providers, utilities and local cloud operators.
“The AI factory will become the bedrock of modern economies across the world,” Nvidia founder and CEO Jensen Huang said in a media Q&A.
Nvidia also outlined what it calls the five components of a national AI strategy: recognizing AI as a national priority, developing an AI-ready workforce, building locally trained AI models and datasets, fostering domestic AI ecosystems, and deploying locally owned AI factories for training and inference.
To illustrate how those principles are being applied, Nvidia pointed to several projects developed through its AI Nations initiative, which the company launched in 2019.
In France, ThinkDeep AI agents built on Nvidia’s platform are helping the Ministry of Economy and Finance automate document-intensive public-service workflows. Nvidia said the system processes millions of documents and data sources, reducing document search times from two days to two minutes, saving 2 million euros for 10,000 employees while lowering energy use through in-country infrastructure.
In India, Nvidia highlighted Sarvam, a platform built on domestic Nvidia GPU infrastructure that delivers multilingual AI models and voice agents across the country’s 22 official languages. The company said the platform enables government and enterprise services to operate in multiple languages while keeping data, computing resources and governance within national borders.
In Brazil, Nvidia said Widelabs is using Nvidia-accelerated infrastructure to support the Public Ministry of Rio Grande do Sul. According to the company, the system streamlines internal investigations and improves access to justice records for more than 8 million citizens across nearly 500 municipalities.
Nvidia said these deployments demonstrate how countries can combine domestic infrastructure, local datasets and homegrown talent to develop AI systems tailored to national priorities while supporting public services and other government workloads.
This analysis is based on reporting from Nvidia.
Image courtesy of Nvidia.
This article was generated with AI assistance and reviewed for accuracy and quality.
