SK Group Chairman Chey Tae-won said on Wednesday that the group's memory chip maker SK Hynix's high-bandwidth memory (HBM) development speed has exceeded Nvidia's required supply speed, suggesting that the bargaining chips with Nvidia have increased.
"Until recently, SK Hynix's HBM development speed was behind Nvidia's requirements, so they urged us to speed up progress," Chey Tae-won said at a press conference at the Las Vegas Convention Center, referring to a meeting with Nvidia CEO Huang Renxun earlier in the day.
"Currently, our development speed is slightly higher than Nvidia (required speed)," he said, "Although this may change, we are developing at a comparable speed now."
SK Hynix is Nvidia's main HBM supplier with the industry-leading fifth-generation HBM3E chip.
The South Korean chip giant announced earlier that the HBM planned for production in 2025 has been sold out, with Nvidia as the largest customer.
"(Huang and I) discussed and confirmed the HBM schedule that was set at the working level, and the supply volume for this year has been determined, but I don't remember the specific number," Chey added.
In addition to HBM, Chey also stressed that SK Group will focus on AI data centers as a major growth driver in the future. This is part of a broader strategy to transform the energy-communications group into a high-tech AI technology company.
"I think energy solutions for AI data centers are crucial," he said, citing key energy issues such as power supply, energy efficiency and cooling. In this regard, the AI data center business is closely related to SK's business portfolio.
