OpenAI is leaning more heavily on South Korea for both the hardware behind AI and the companies using it.
The ChatGPT maker is expanding its work with Samsung Electronics around next-generation chip research and production as demand for advanced memory climbs. At the same time, ChatGPT Enterprise use is growing rapidly among Korean businesses. For local distributors, integrators, and service providers, those developments put new infrastructure spending and enterprise AI projects in the same market.
Partners do not need to wait for another OpenAI chip to see where opportunities may emerge. Samsung and SK Hynix are already tied to OpenAI’s infrastructure plans, while Korean companies are putting ChatGPT into more parts of their operations.
Samsung moves closer to OpenAI’s chip plans
Harrison Kim, general manager of OpenAI Korea, said the companies have made progress on joint chip work.
“One of the areas where we have made the most progress and gained the most recognition with Samsung Electronics is our joint production and research on the next-generation chips we are developing,” Kim said, according to Reuters.
OpenAI Korea later clarified that the comments referred to its existing relationship with Samsung and said there was “nothing new to announce.” Samsung also declined to confirm information involving customers.
SamMobile reported that Samsung will manufacture a custom OpenAI chip. OpenAI, however, has not publicly confirmed Samsung as the manufacturer of its next processor.
OpenAI’s first custom inference processor, Jalapeno, was developed with Broadcom and will be manufactured by TSMC in Taiwan.
The setup already spreads OpenAI’s chip relationships across major APAC suppliers. Taiwan handles fabrication for Jalapeno, while South Korean companies are becoming increasingly important for memory and future chip work.
Memory demand puts Korean suppliers in focus
Samsung and SK Hynix signed letters of intent last year to supply memory chips for OpenAI’s Stargate data center project. Kim said OpenAI expects memory demand to keep growing as AI workloads require faster and more complex computing.
Samsung is preparing for that demand. Quartz said that the company has shipped samples of its HBM4E high-bandwidth memory to major global customers, including Nvidia, AMD, and Google. The memory is designed for AI accelerators such as Nvidia’s Rubin and Google’s Ironwood.
For infrastructure partners, the relevant question is what gets built around those processors and memory systems. More AI deployments can mean projects involving servers, networking, storage, cooling, and data center capacity, areas where distributors and systems integrators already participate.
ChatGPT growth opens another lane for partners
Samsung is also one of OpenAI’s largest ChatGPT deployments globally. Employees in Korea and overseas use the platform across research and development, marketing, and sales, according to OpenAI.
SamMobile noted that Samsung’s Device eXperience division permits employees to use ChatGPT alongside Google Gemini and Anthropic Claude, giving staff access to several competing AI platforms.
OpenAI’s growth in South Korea is not limited to Samsung. According to Reuters, the company said ChatGPT Enterprise usage among South Korean companies and institutions grew roughly 28-fold year over year by the end of August.
Korean MSPs and integrators may find more work around connecting AI to business systems, setting access controls, managing company data, and supporting multiple AI platforms as deployments grow.
OpenAI has not confirmed what role Samsung will play in its next custom processor. Channel firms already have something more concrete to watch, though: South Korea is becoming a larger source of the memory OpenAI needs and a fast-growing market for the enterprise AI services built on top of it.
Also read: Samsung and Arm are reportedly developing a 2nm AI chip as Samsung expands its custom-silicon business.