Sam Altman meets with Samsung leadership to secure more RAM for data centers amid crushing shortages — OpenAI CEO travels to South Korea to meet Samsung leadership to discuss future AI data centers

Altman's Seoul trip is fundamentally a "capacity sovereignty" negotiation. With Samsung and SK hynix controlling 83% of HBM output, OpenAI has shifted from buyer to hostage of a duopoly's pricing power. HBM is scarcer than GPUs—a memory bandwidth outage halts training pipelines instantly. On the technical front, the HBM4 transition window (2H25–2026) represents Samsung's "lane-change" opportunity. Trailing SK hynix by roughly 6–9 months on HBM3E yields, Samsung faces a fresh thermal-management gauntlet with 12-layer TSV stacking. Securing a long-term OpenAI contract could lift HBM margins from single digits to 15%+. The real wildcard is Micron. As the sole US-based HBM supplier, it carries a structural "compliance premium" as export controls tighten. The 83% Korean share could compress rapidly under geopolitical stress. Over the next 12–24 months, CXL memory pooling and optical interconnects will gradually erode HBM's irreplaceability. But until then, memory suppliers' bargaining power will only intensify. Altman didn't fly to Seoul to buy DRAM—he went to lock in two years of production priority.