Samsung has reportedly flagged deeper stress in the global memory market, warning that supply constraints could intensify over the next few years as demand, particularly from AI-driven use cases, continues to accelerate. According to a report by 9To5Google, speaking during its latest earnings call, the company indicated that the current mismatch between supply and demand is already significant, and may worsen further, with early indicators suggesting that the situation could become more pronounced by 2027.


Memory shortage may worsen by 2027

As per the report, during the earnings call, Samsung explained that demand for memory products is being increasingly driven by AI-related workloads, which continue to scale rapidly across industries. However, on the supply side, the company highlighted structural limitations. Expanding the production volume of DRAM (Dynamic RAM) chips is a time-intensive process, and new semiconductor factories cannot do it quickly enough to immediately address the surge in demand. 

 


The company said inventory levels remain extremely tight at present, with available supply falling well short of customer requirements. It also noted that its demand fulfilment rate has dropped to a record low. In a shift from previous cycles, customers are now placing orders much earlier than usual, with some already securing supply for 2027 amid concerns of prolonged shortages. 


Based on its current order pipeline and pre-booked demand, Samsung expects the gap between supply and demand to widen further over the next two years, signalling continued pressure on pricing and availability across the broader tech ecosystem.


Focus on efficiency, AI-led devices


On the mobile side, Samsung said it is prioritising cost efficiency across its smartphone portfolio as industry-wide cost pressures remain elevated. At the same time, the company reiterated its strategy of driving flagship growth through AI integration and hardware innovation. 


It also offered an early glimpse into upcoming product categories, including AI glasses aimed at delivering immersive, multimodal experiences. Alongside this, Samsung confirmed that it is continuing to strengthen its foldable lineup, with future iterations expected to build on both form factor and feature enhancements. 


Looking ahead, the company acknowledged that 2026 is likely to remain a challenging year, citing geopolitical uncertainties and rising input costs. However, it maintained that its focus on AI-led differentiation and operational efficiency will be key to navigating these headwinds while sustaining growth in its premium segment.


Samsung flags possible losses in phone business


Earlier this week, Samsung MX (mobile experience) head TM Roh warned the company’s leadership that it could see its smartphone business slip into losses for the first time, as rising DRAM and NAND prices significantly push up production costs. 


Despite steady demand and strong sales of its flagship Galaxy S26 series, an Ars Technica report citing comments from TM Roh suggests that the mobile division is under growing margin pressure, potentially reversing its long-standing profitability even during past economic downturns. 


This statement came on the heels of data from Counterpoint Research suggesting memory costs are set to take up a larger share of smartphones’ price. As per the report, by mid-2026, DRAM and NAND could account for over one-third of the total cost of an entry-level smartphone, while in premium devices, memory is expected to contribute roughly one-fifth of overall costs.


Why memory supply remains tight


The spike in memory prices is tied to a broader global shortage of DRAM and NAND chips. As per a report by Nikkei Asia, supply constraints are expected to persist until around 2027. Even with capacity expansion plans in place, global DRAM output may only be able to meet about 60 per cent of total demand, leaving a potential shortfall of up to 40 per cent in certain scenarios. 


One of the key reasons behind this imbalance is the growing industry shift towards high-bandwidth memory (HBM), which is increasingly used in AI infrastructure and offers better margins. This transition has redirected production capacity away from conventional memory used in smartphones and PCs. 


Additionally, new fabrication units will take time to ramp up to full-scale production, delaying any meaningful easing of supply pressures. Meanwhile, memory prices in early 2026 are estimated to have surged by nearly 90 per cent on a quarter-on-quarter basis, further driving up costs for device manufacturers.



Source link

YouTube
Instagram
WhatsApp