Samsung memory chief, Nvidia CEO discuss next-generation HBM cooperation
Jun Young-hyun, right, head of Samsung Electronics' device solutions division, poses with Jensen Huang, CEO of Nvidia, at a hotel in Seoul, Monday. Courtesy of Samsung Electronics
The head of Samsung Electronics' memory business said Monday he met with Nvidia CEO Jensen Huang and discussed long-term cooperation in next-generation high-bandwidth memory (HBM) products.
Jun Young-hyun, who leads Samsung Electronics' device solutions division, made the remarks to reporters after meeting Huang during his visit to Korea to meet industry leaders to explore broader cooperation opportunities across the artificial intelligence (AI) ecosystem.
"We have maintained cooperation with CEO Huang for a long period, and we have had very productive discussions so far," Jun said.
A key focus of the talks was long-term collaboration on next-generation HBM technologies, including HBM4E and HBM5, as well as cooperation in the foundry business, Jun said.
HBM chips are advanced memory semiconductors that work alongside AI processors to provide the massive bandwidth needed for AI training and inference, making them a critical component of modern AI systems.
Jun added that in the shorter term, Samsung Electronics will make efforts to supply sufficient HBM4 products to Nvidia, which accounts for a dominant share of the global AI accelerator market.
Their meeting came as Samsung seeks to strengthen its position in the rapidly growing HBM market, where demand is being driven by the expansion of AI data centers and the deployment of increasingly powerful AI models.
Responding to a question about Huang's previous remark that SK hynix, Samsung's Korean rival, will remain Nvidia's biggest memory partner, Jun replied that the company will "work hard on what we do" and prove it with results.