Could a gigawatt-scale AI cloud reshape South Korea’s technology landscape? SK Telecom and NVIDIA are joining forces to expand AI computing capacity for industry.
SK Telecom and NVIDIA have announced plans to develop a gigawatt-scale AI cloud infrastructure in South Korea, with the first AI factory expected to become operational in 2027 as part of efforts to expand the country’s AI computing capacity.
The project will be built using NVIDIA’s DSX platform and is intended to support a range of AI applications, including sovereign AI models, enterprise AI systems, physical AI technologies and agentic AI services. The infrastructure is expected to serve businesses and industries across South Korea, with potential expansion into other Asian markets.
According to the companies, the AI cloud will comprise multiple AI factories designed specifically for AI training, inference and autonomous AI workloads.
Unlike conventional cloud platforms that support a broad range of computing needs, the proposed infrastructure will focus on graphics processing unit (GPU)-based computing tailored to AI development and deployment.
The initiative builds on South Korea’s established presence in sectors such as telecommunications, semiconductors, memory technologies, manufacturing, robotics and mobility, where AI adoption is increasingly moving into large-scale commercial operations.
SK Telecom said the project would strengthen its AI infrastructure capabilities across hardware, data centre operations and network services. The company has recently expanded its AI activities, including work on digital twin technology for semiconductor manufacturing facilities and participation in South Korea’s Sovereign AI Foundation Model Project.
As part of the collaboration, SK Telecom will also join NVIDIA’s Cloud Partner programme, enabling it to offer AI cloud services based on NVIDIA’s latest computing infrastructure and software ecosystem.
In addition to deploying the new infrastructure, NVIDIA and SK Group plan to undertake joint research into future AI factory architectures. The collaboration will focus on improving computing performance, memory technologies, energy efficiency and data centre operations to support the next generation of AI infrastructure.
The companies said they will also explore opportunities to optimise AI factory operations and improve the scalability and resilience of AI services.
