Samsung partners with OpenAI on next-gen chips

Samsung Electronics is deepening its partnership with OpenAI, focusing on next-generation chip development while simultaneously rolling out ChatGPT across its internal operations. The collaboration signals a significant shift in how the South Korean tech giant approaches artificial intelligence infrastructure, combining hardware advancements with widespread software deployment.

Chip Development and Memory Supply

Harrison Kim, general manager of OpenAI Korea, highlighted joint production and research on next-generation chips as a cornerstone of the expanding relationship. Kim noted that the companies have made the most progress in this specific area, though he did not disclose details about the chip’s specifications or Samsung’s exact manufacturing role. This development follows OpenAI’s June announcement of Jalapeno, its first custom AI chip, which was designed with Broadcom and is slated for manufacturing by TSMC.

OpenAI’s interest in Samsung extends beyond chip design to include critical memory components. Large language models require substantial memory capacity and bandwidth to function efficiently in data centers, a need that Kim says will only grow as computing becomes faster and more complex. Last year, Samsung and SK Hynix signed letters of intent to supply memory chips for OpenAI’s Stargate data center project, with Kim emphasizing that working with South Korean semiconductor suppliers remains a priority for the local office.

The reliance on advanced memory solutions creates a natural dependency between the companies. Samsung’s manufacturing capabilities and OpenAI’s software expertise appear to be moving in parallel, suggesting a long-term strategic alignment rather than a one-off project. This integration could eventually reduce latency and improve efficiency for OpenAI’s infrastructure as it scales.

Internal Deployment and Enterprise Growth

While the hardware side of the partnership focuses on future capabilities, Samsung is currently integrating OpenAI’s software into its existing workforce. Kim described the deployment as “one of the largest-scale deployments of ChatGPT globally,” with employees using the tool for research, development, marketing, and sales. Samsung has been expanding internal access to generative AI tools, recently allowing staff in its Device eXperience division to use ChatGPT, Google’s Gemini, and Anthropic’s Claude.

The adoption rate in South Korea has surged. OpenAI reported that the number of Enterprise users in the country increased roughly 28-fold by the end of August compared to a year earlier. The company noted that businesses are applying AI across various departments—legal, finance, operations, and sales—rather than restricting its use to isolated tasks. This widespread integration suggests that Samsung is not merely testing the waters but is embedding AI deeply into its corporate processes.

As OpenAI prepares to release its GPT-6 Astra model, the focus is shifting toward applications that require more subtle interaction, such as professional work, coding, and cybersecurity. The success of these advanced models will likely depend on the very memory and infrastructure partnerships currently being solidified with Samsung and other suppliers. The rapid expansion of ChatGPT Enterprise users indicates that enterprises are preparing for a future where these systems are standard tools for daily operations.