OpenAI Expands Its Semiconductor Strategy

OpenAI is deepening its relationship with Samsung Electronics as the artificial-intelligence company moves further into custom semiconductor development. Harrison Kim, General Manager of OpenAI Korea, said the companies have made significant progress in joint production and research involving next-generation chips, although he provided few technical details about the project.

The development signals a broader shift in how major AI companies are approaching computing infrastructure. Rather than relying exclusively on external chip designers and manufacturers, OpenAI is increasingly developing specialized hardware to support the enormous computational requirements of its AI services. In June, the company unveiled Jalapeno, its first custom AI chip developed with Broadcom, designed primarily for AI inference workloads. TSMC is expected to manufacture those chips.

Samsung’s Role Extends Beyond Advanced Chips

The relationship with Samsung is significant because it potentially spans multiple layers of the AI semiconductor supply chain. Samsung Electronics and SK Hynix signed letters of intent last year to supply memory chips for OpenAI’s Stargate data-center project, placing South Korean semiconductor manufacturers at the center of the company’s infrastructure ambitions.

OpenAI expects demand for advanced memory to continue increasing as AI systems require faster and more complex computing. That makes cooperation with Korean semiconductor suppliers strategically important, particularly as AI infrastructure expands from model training toward large-scale inference and enterprise deployment.

For Samsung, deeper cooperation with OpenAI could provide exposure to one of the world’s fastest-growing sources of semiconductor demand. However, Samsung itself is simultaneously expanding internal use of generative AI, with employees in its Device eXperience division permitted to use ChatGPT, Google’s Gemini and Anthropic’s Claude for internal work.

South Korea Is Becoming a Major AI Enterprise Market

The semiconductor relationship is developing alongside rapid growth in OpenAI’s commercial footprint in South Korea. The company said ChatGPT Enterprise users at South Korean companies and institutions increased approximately 28-fold by the end of August compared with a year earlier.

That expansion suggests businesses are moving beyond limited individual experiments and integrating AI into broader organizational functions, including legal, sales, finance, marketing and operations. OpenAI also said South Korea had the largest number of paying ChatGPT subscribers outside the United States as of late last year, underlining the market’s importance to its international strategy.

AI Adoption Could Strengthen the Semiconductor Cycle

OpenAI expects enterprise adoption to accelerate following the September 4 launch of GPT-6 Astra, which the company says improves performance across computer use, professional work, science, coding and cybersecurity. If increasingly capable models drive greater enterprise usage, demand for computing capacity and supporting memory infrastructure could expand accordingly.

That creates a reinforcing relationship between software adoption and semiconductor demand. More sophisticated AI applications require greater computing resources, while better hardware enables broader deployment. For investors in the U.S. and Israel, the OpenAI-Samsung relationship therefore represents more than a corporate partnership: it offers another indication that AI competition is increasingly moving into the underlying infrastructure that powers the industry.

The key questions ahead will be the technical scope of OpenAI’s next-generation chips, Samsung’s eventual role in manufacturing or supplying components, and the scale at which OpenAI can convert enterprise adoption into sustained computing demand. As AI infrastructure spending expands, partnerships that combine chip design, manufacturing and memory capacity could become increasingly important to the industry’s competitive landscape.