Sam Altman meets Samsung leadership to secure more RAM for AI data centers
The meeting follows a surge in demand for memory chips as AI data centers expand. Samsung and SK Hynix control 83% of HBM revenue, intensifying competition for memory chip supplies.

Sam Altman, CEO of OpenAI, has met with Samsung leadership in South Korea to address shortages of RAM for AI data centers. The discussions focus on securing additional memory chip supplies as demand for high-performance computing grows rapidly. This move highlights the critical role of memory chips in advancing AI capabilities and the increasing pressure on global semiconductor manufacturers.
The meeting took place amid a global shortage of high-bandwidth memory (HBM) chips, which are essential for training large AI models. Samsung and SK Hynix currently control approximately 83% of the HBM market, giving them significant leverage in negotiations with AI companies. This dominance has raised concerns about potential bottlenecks in the supply chain and the concentration of power among a few key players.
Samsung and SK Hynix's combined control over 83% of HBM revenue underscores their central role in the AI memory chip market. This market concentration has intensified competition for memory chips as AI data centers scale up operations. With demand for HBM expected to grow significantly in the coming years, the ability to secure supply will be a key factor in the success of AI companies.
The reliance on a limited number of suppliers for critical components like HBM raises concerns about cost, vendor lock-in, and governance. Companies may face higher costs and reduced flexibility if they depend on a small group of manufacturers. Additionally, the concentration of power could lead to governance challenges, as suppliers may influence the direction of AI development through control over essential resources.
As the meeting between Sam Altman and Samsung leadership continues to unfold, the outcome will have broader implications for the AI industry. The availability of memory chips will remain a critical factor in the pace of AI innovation, and the balance of power among semiconductor manufacturers will shape the future of AI development.