Nvidia and SK Hynix have signed a multi year technology partnership to develop next generation memory for AI factories, deepening one of the most important supply relationships in the global chip industry.

The agreement focuses on advanced memory technologies, including high bandwidth memory used in Nvidia’s AI systems. The deal comes as demand for AI infrastructure continues to outpace supply across memory, wafers, packaging, and other critical parts of the semiconductor supply chain.

Nvidia CEO Jensen Huang said the memory shortage is likely to last for years, underscoring how tight supply has become as cloud providers, governments, and large companies race to build AI data centers.

SK Hynix is already Nvidia’s largest memory partner and a key supplier of HBM chips used in its AI accelerators. The new agreement formalizes that relationship around future platforms, shifting the partnership beyond component supply and into deeper technology development.

The deal also comes as SK Hynix prepares a major capacity expansion. The company plans to double wafer production capacity over the next five years to meet AI driven demand, according to SK Group Chairman Chey Tae won.

For Nvidia, the agreement helps secure a critical part of its AI hardware roadmap. For SK Hynix, it strengthens its position in the most profitable corner of the memory market at a time when AI demand is reshaping the semiconductor cycle.