Silicon Motion's CEO, Chia-Chang Gou, stated that memory (DRAM) and SSD (NAND) shortages will persist until 2028, a longer timeline than previously expected. The primary driver is the relentless demand from AI firms, which are securing supply through long-term contracts and prepayments. Current production capacity only meets 60-70% of total demand, and new factories won't achieve volume production until late 2027 or early 2028. Prices are expected to rise in the second half of 2024, forcing discontinuation of some consumer products. The shift from AI training to inference is further accelerating demand. This situation benefits memory manufacturers and controller makers but pressures downstream consumers and enterprises.
The timeline for memory and SSD shortages has been extended from 2027 to 2028, and AI firms are now actively securing supply via long-term contracts, exacerbating the supply-demand gap.
Unchanged: The fundamental supply-demand imbalance caused by AI-driven demand remains; memory makers still face capacity expansion delays; prices continue to rise.
Cautious, as the news signals prolonged supply tightness and rising costs, benefiting memory/controller makers but harming consumers, enterprises, and smaller AI firms.
AI firms benefit from priority supply but face higher costs; the shortage may slow AI deployment for some.
Beneficial for memory and controller makers, negative for downstream hardware vendors and consumers.
Creates opportunities for suppliers but risks for buyers; strategic planning becomes critical.
Cloud providers face higher memory/storage costs and potential capacity constraints for AI workloads.
As a controller maker, it benefits from increased demand and higher prices.
Inundated with customer prepayments; benefits from long-term contracts and price rises.
Expects severe shortages; positioned to gain from high memory prices.
Secure supply but at higher cost; smaller firms may be disadvantaged.
Face higher prices for electronics and potential product discontinuations.
The extension of the timeline indicates the AI boom is structurally reshaping the memory and storage industry, creating persistent supply constraints that will affect pricing, product availability, and innovation for years. Companies planning hardware deployments must factor in elevated costs and longer lead times. This also signals that the shift from AI training to inference may be accelerating, further straining supply.
Higher component prices and discontinued products lead to increased costs for electronics and limited availability.
Datacenter expansion faces memory/storage constraints and higher costs, impacting infrastructure plans.
They secure supply but at higher prices, and smaller firms may struggle to compete for contracts.
They benefit from higher prices and long-term contracts, with strong demand outlook.
Elevated hardware costs increase capital requirements and may slow down AI deployment.
Positive for memory makers worldwide, negative for global consumers and enterprises due to higher costs.
AI firms (e.g., NVIDIA) may secure supply but face higher costs; consumers see price increases.
Domestic demand pressures local manufacturers, but they also face export restrictions; consumers face shortages.
No direct cybersecurity implications.
No direct data governance implications.
No reputational damage implied for major entities.
Factory build schedules may be delayed, further extending shortages.
Datacenter expansion and AI infrastructure deployment are directly hindered by memory/storage shortages.
Trade tensions and export controls could exacerbate supply constraints, especially for Chinese players.
No direct regulatory changes implied.
Long lead times and insufficient capacity create critical supply chain bottlenecks.
No direct workforce impact.
No AI liability issues raised.
Constrained access to memory/storage and rising costs for expansion.