Samsung's latest earnings call shows that artificial intelligence demand is broadening across multiple types of memory products, not concentrated in just the most advanced chips.
The company expects memory supply to remain tight through 2028, and long-term contracts are now the mechanism reshaping how memory is allocated and supplied to customers.
What happened
Samsung's Q2 2026 earnings call revealed that AI is expanding demand across a broader range of memory products, not just the highest-end chips. Memory was about one-third of the company's Q&A discussion, and the company indicated supply tightness will persist through 2028.
Why it matters
Memory constraints have been a bottleneck for AI infrastructure. Samsung's signal that demand is diversifying into a wider product mix—and that shortages will continue for over two years—suggests AI adoption is moving beyond a few flagship applications into more varied deployments, reshaping how enterprises must plan their infrastructure investments.
What to watch
Long-term agreements (LTAs) are reshaping how Samsung and other suppliers allocate memory. This shift in supply contracting could determine which customers secure reliable access and how pricing and capacity planning evolve through 2028.
Samsung Electronics' second-quarter 2026 earnings call highlighted a meaningful shift in how artificial intelligence is driving demand for semiconductor memory. According to DIGITIMES Intelligence's analysis of the call, memory dominated the discussion, occupying roughly one-third of the Q&A session—a proportion that underscores how central memory capacity has become to Samsung's AI narrative. The three key insights from the call reveal a maturing AI infrastructure market. First, AI demand is no longer confined to the absolute highest-end memory products; instead, it is spreading into a broader range of memory SKUs (stock-keeping units), indicating that AI workloads are diversifying beyond headline applications like large language model training. This could reflect the emergence of smaller, specialized AI models, enterprise-grade inference systems, and edge AI deployments that do not require the most expensive premium memory but still depend on reliable capacity. Second, Samsung indicated that memory supply will remain constrained through 2028—a two-year-plus window that suggests the market is not expecting a near-term relief from capacity pressures. This extended outlook implies that capital investment in memory manufacturing will need to keep pace with broadening AI demand, and customers cannot expect prices to fall sharply or availability to loosen materially in the near term. Third, long-term agreements (LTAs) are fundamentally reshaping how memory supply is allocated and how supplier operations are structured. Rather than relying on spot markets or short-term purchase orders, manufacturers and customers are locking in multi-year contracts, allowing suppliers to forecast demand with greater certainty and customers to secure predictable access. This contractual shift is significant because it moves the memory market away from transactional purchasing toward partnership-based supply relationships, altering the competitive dynamics and margin structures for both suppliers and their customers.
Samsung's Q2 2026 earnings call underscores a critical shift in AI infrastructure: demand for memory is no longer concentrated in a single ultra-premium product category. Instead, the diversification into a wider product mix signals that AI workloads are becoming more varied—moving beyond large language model training and inference into smaller models, edge deployments, and specialized tasks. This broadening is reflected in the fact that memory accounted for about one-third of the earnings call's Q&A, indicating investor and analyst focus on the memory business as a key indicator of AI momentum. The persistence of supply tightness through 2028 suggests that this diversification will not quickly ease capacity constraints; rather, demand across multiple memory tiers is sustaining pressure on production. Long-term agreements emerging as a supply-shaping mechanism indicates that both suppliers and customers are moving away from spot market purchases toward contracted relationships, allowing manufacturers to plan capacity allocation and customers to secure predictable access—a structural shift that will likely influence pricing, margin dynamics, and competitive positioning for the next two years.
AI-summarized, only the topics you pick — one digest a day via Email, Slack, or Discord.
Free · takes 30 seconds · unsubscribe anytime
Ask AI anything about this article. Q&As are published on this page for other readers too.
The AI news that matters, in one minute each morning.
Sign up free