Samsung in Talks with Nvidia to Supply Next-Gen HBM4 Chips
Samsung Electronics has announced it is currently in talks with Nvidia to supply next-generation high-bandwidth memory (HBM4) chips. The South Korean company aims to catch up with competitors in the rapidly evolving AI chip market.
Samsung’s HBM4 Development Plans
The planned release of HBM4 chips is set for next year. However, Samsung has not disclosed a specific shipping date for these advanced memory chips. As the demand for AI technologies escalates, these chips are considered critical for AI chipsets.
Competition in the AI Chip Market
- Local rival SK Hynix has announced plans to begin shipping its own HBM4 chips in the fourth quarter of this year.
- Nvidia remains the top customer of SK Hynix for HBM chips.
- Samsung is also supplying its current-generation HBM3E chips to various clients, including Nvidia.
Analysts assert that the introduction of HBM4 chips will be crucial for Samsung’s strategy to reclaim its competitive edge in the memory market. This follows last year’s restructuring of its chip division due to underperformance in the AI-focused sector.
Market Dynamics and Future Outlook
Samsung’s latest earnings have shown a recovery, attributed to a rise in demand for conventional memory chips. Nonetheless, the company has been slower to adapt to the AI-driven memory chip boom compared to its competitors.
HBM technology, which involves vertically stacking chips, was first introduced in 2013. This design allows for more efficient processing of large datasets in complex AI applications.
As Samsung pushes forward with its HBM4 production, the cooperation with Nvidia signals a pivotal moment for the South Korean semiconductor leader. The successful launch of HBM4 is anticipated to significantly enhance its standing in the competitive AI memory space.
