Samsung Electronics said on October 29, 2026, that AI-driven demand was likely to tighten memory-supply shortages further in 2027 than in calendar year 2026. Its Q3 outlook links that forecast to growing investment in AI infrastructure and demand for memory and storage used in servers. Samsung’s Q3 2026 presentation describes a company forecast, not a measured future outcome.
What Samsung expects from memory supply in 2027
Samsung expects AI-infrastructure investment to accelerate as the AI ecosystem expands. It also anticipates further memory demand around products for new CPU and GPU platforms.
The products named across the presentation’s results and outlook include HBM, or high-bandwidth memory; server DRAM, or dynamic random-access memory; server SSDs; high-speed server DDR5; and QLC SSDs. That list centers the outlook on AI- and server-related demand.
What Samsung’s Q3 figures measure
Samsung reported these Q3 2026 results for its semiconductor segment and consolidated company:
| Reporting scope | Q3 2026 revenue | Q3 2026 operating profit |
| Semiconductor segment | KRW 147.1 trillion | KRW 106.5 trillion |
| Consolidated company | KRW 195.3 trillion | KRW 107.4 trillion |
Samsung’s presentation describes the semiconductor segment’s operating profit as 72% of that segment’s revenue. It says the Q3 financial information was issued before the external review was complete, and that the review could change some figures.
Server demand led reported memory-price increases
Samsung said memory-market prices had increased across applications, led by strong server demand. Its statement describes market prices, rather than a forecast for consumer retail prices.
Intel CEO Lip-Bu Tan separately warned about memory supply in 2027. NeoTeo covered that statement in its earlier report on Intel’s warning.
