Global Telcos Embrace NVIDIA-Powered AI Infrastructure to Drive Sovereign Innovation Across Five Continents

Telecom companies around the world are rapidly establishing NVIDIA-powered sovereign AI infrastructure, marking a significant shift towards localized, high-performance AI cloud services. Over the past 18 months, 18 telcos across five continents—North America, South America, Europe, Asia, and Africa—have launched AI factories, with many more planning to follow suit. This move is driven by the recognition that artificial intelligence (AI) is crucial for economic growth and innovation, making it a national priority for many countries. Key Developments in Sovereign AI Infrastructure Sovereign AI: A National Imperative The concept of "sovereign AI" refers to the infrastructure that allows countries to develop and control AI models and data within their borders. This ensures that AI technologies align with local cultural, ethical, and economic priorities, and comply with regional security policies and regulations. The compute divide is becoming the new digital divide, emphasizing the need for nations to invest in their own AI supercomputing systems. Telecom Companies: Trusted Providers Telecom operators are uniquely suited for this role due to their existing extensive data networks, robust cybersecurity measures, and deep understanding of localized needs. They can leverage their infrastructure and expertise to offer GPU-accelerated AI services, which are essential for the training, fine-tuning, and inferencing of AI models. Telcos can also ensure data sovereignty and compliance with local laws, a critical factor in gaining trust and supporting national initiatives. Global Examples of Telco-Powered AI Factories Japan SoftBank Corp. and KDDI are building large-scale AI data centers and developing Japanese-language large language models (LLMs) and enterprise AI solutions. These investments aim to boost local AI innovation and address regional challenges. Southeast Asia Indosat Ooredoo Hutchison in Indonesia is offering AI services for citizen services, healthcare, and localized open-source LLMs. Singtel's RE:AI in Singapore supports the National AI Strategy 2.0, providing AI cloud services to enterprises, government agencies, researchers, and academia. YTL AI Cloud in Malaysia supports the country's digital economy ambitions. FPT has launched AI factories in Vietnam and Japan, offering NVIDIA GPU-accelerated computing platforms and a suite of ready-to-use AI applications. India Tata Communications is deploying AI cloud infrastructures for sectors like manufacturing, healthcare, retail, and financial services, enabling localized AI development and adoption while maintaining data sovereignty. Middle East Ooredoo is investing in AI factories in Qatar, Algeria, Kuwait, the Maldives, Oman, and Tunisia, providing GPU-as-a-Service (GPUaaS) and localized AI ecosystem support. Europe Scaleway in France is expanding its AI cloud capacity with high-end GPUs to support startups and enterprises. Telenor in the Nordics has launched a sovereign, secure, and sustainable AI factory for both internal and external use. Swisscom is building the Swiss AI platform, featuring the country’s first NVIDIA SuperPOD. Fastweb in Italy has launched the NeXXt AI factory, powering the Italian AI Model (MIIA) and offering AI services to various sectors. Latin America Telconet in Ecuador has deployed an AI factory to enhance urban safety, support multiple use cases, and introduce the latest-generation GPUs to the region. North America TELUS is the first North American service provider to join NVIDIA’s Cloud Partner program, aiming to bring a comprehensive sovereign AI factory to Canada. The platform will be housed in sustainable data centers powered by 99% renewable energy, set to launch this summer in Quebec. TELUS is focusing on delivering secure, high-performance AI services to Canadian businesses, startups, and researchers. Africa Cassava Technologies is pioneering the first pan-African NVIDIA Cloud Partner (NCP) initiative, starting with a deployment in South Africa and expanding to Egypt, Kenya, Morocco, and Nigeria. Cassava is working with African governments to drive national AI development, including the creation of localized AI models and applications. Central Asia Kazakhtelecom in Kazakhstan is launching a commercial AI factory powered by NVIDIA GPUs and AI Enterprise. The platform will cater to public institutions, private businesses, startups, and universities, supporting key sectors like healthcare, banking, and energy. Diverse Users and Applications As telco AI factories mature, they are serving a wide range of users and applications: - Nations: Developing national LLMs, AI-powered citizen services, and supporting AI education. - Enterprises: Building custom AI models and integrating agentic AI into workflows, particularly in healthcare, manufacturing, financial services, and retail. - Model Builders: Startups and research institutions creating and testing new AI models. - Internal Telcos: Enhancing customer service and network operations with AI, including digital humans and field operations AI agents. Future Trends: Scaling Up and Out The next frontier for telcos is to build an AI grid at a national scale. This involves: - Scaling Up: Expanding centralized AI factories for compute-intensive tasks like training and reasoning. - Scaling Out: Deploying distributed AI-RAN (Radio Access Network) data centers for edge AI workloads, including inferencing. NVIDIA provides a cohesive reference architecture and capabilities like NVIDIA Cloud Functions, which help orchestrate demand and supply across the AI grid and AI-RAN. Industry Insights and Company Profiles McKinsey Research suggests that the global GPUaaS market addressable by telcos could reach $35 billion to $70 billion annually by 2030. NVIDIA’s Cloud Partner Program (NCP) offers a comprehensive reference architecture that includes a full-stack technology platform, demand generation support, and education for upskilling. This program is instrumental in enabling telcos to establish themselves as AI service providers quickly and efficiently. SoftBank Corp. and IOH are leading examples in building AI grids, demonstrating how telcos can integrate AI factories and AI-RAN to create, distribute, and consume intelligence nationally. By leveraging NVIDIA’s NCP and building AI grids, telecom companies are not only extending their role as connectivity providers but also becoming pivotal players in the AI economy. They are transforming into platforms that foster inclusive, sovereign AI development, driving a new wave of innovation and productivity worldwide.