The memory boom can continue while AI data center demand keeps outrunning available DRAM, HBM, and NAND supply, but the supplied brief does not prove that the boom is structurally permanent. The strongest evidence in the brief is price-driven: DRAM spot prices and NAND wafer prices are described as rising about tenfold, while hyperscaler capital expenditure is expected to rise sharply through 2026. That supports a near-term boom narrative, but readers should treat it as a high-volatility cycle until supply expansion, demand durability, and price normalization are checked against newer data.
| Primary source | Wallstreetcn |
|---|---|
| Reported at | 2026-07-14T14:37:10.000Z |
| Topic | AI Crypto |
| Evidence limit | Reported facts are separated from interpretation; current prices and platform terms require independent verification. |
Evaluate OKX for your use case
Check regional eligibility, current fees and product availability on the official destination.
Review OKXDirect Market Reading
The supplied brief argues that memory has moved beyond a normal semiconductor upcycle. It cites WSTS-based data showing MOS memory monthly shipments near 5.6 billion dollars in 2016, around 5.8 billion dollars at the early-2023 trough, and about 63.3 billion dollars by May 2026.
The brief also says recent memory year-over-year growth reached 285%, compared with roughly 60% during the previous memory bubble around 2017. On the supplied numbers alone, the current cycle is presented as historically unusual, not just strong.
Why Prices Matter
The brief's central mechanism is pricing. It says DRAM spot prices rose from 4.70 dollars in early 2025 to 46.00 dollars recently, while NAND wafer prices rose from 2.40 dollars to 25.00 dollars. That means market value can surge even if physical shipments do not rise at the same pace.
This matters for analysis because price-led revenue growth can reverse faster than demand-led structural growth. A durable boom needs sustained demand, constrained supply, and disciplined capacity allocation. The supplied brief supports the first two conditions, but it does not fully establish how long producers can maintain them.
AI Data Center Demand
The supplied event links the memory shortage to heavy AI data center spending by Amazon, Google, Microsoft, and Meta. It states that their combined capital expenditure was 21 billion dollars in 2015, is expected to reach 355 billion dollars in 2025, and is expected to reach 755 billion dollars in 2026.
The brief's explanation is that AI training and inference pull in GPUs, HBM, high-performance DRAM, and NAND-backed SSDs. Memory makers then prioritize higher-margin AI and data center products, reducing capacity available for PCs, smartphones, and game consoles.
Crypto Relevance
For an OKX-oriented crypto audience, the link is indirect. AI infrastructure spending can influence narratives around compute, data centers, storage, energy use, hardware supply chains, and companies tied to AI infrastructure. The supplied brief does not name any crypto asset, protocol, exchange metric, or on-chain indicator.
That means the article should not be used as a token-price forecast. It is more useful as a macro context piece: if traders are watching AI-related crypto narratives, they should also watch semiconductor supply, memory pricing, hyperscaler capital expenditure, and whether consumer-device shortages start affecting broader technology sentiment.
Evidence Limits
This analysis uses only the supplied event and brief. It does not independently verify WSTS, TrendForce, company capital expenditure, or later market updates. Some figures in the brief are forecasts, including 2026 and 2027 semiconductor market estimates, so they should be treated as forward-looking rather than settled outcomes.
The supplied brief also emphasizes the author's interpretation that prior semiconductor forecasts were too conservative. That is useful context, but it is still one event source. Before making business, investment, or procurement decisions, readers should compare the claims with current primary data and company filings.
Practical Checks
The most practical check is whether DRAM and NAND prices are still rising, stabilizing, or reversing. If prices flatten while capacity improves, the revenue boom could cool even if AI demand remains strong.
A second check is whether hyperscaler capital spending remains near the trajectory described in the brief. A third check is whether memory makers keep prioritizing HBM and data center-grade products over consumer-market supply. A fourth check is whether PC and smartphone makers continue reporting memory shortages and cost pass-through pressure.
Risk Disclosure
This is not financial advice and does not recommend buying, selling, or holding any crypto asset, equity, or semiconductor exposure. The supplied brief supports an AI-driven memory-boom thesis, but it does not remove cycle risk, forecast risk, supply-response risk, or valuation risk.
For readers using OKX to follow AI and crypto-market narratives, the supplied OKX link and code can be treated only as an access path: OKX official destination with code 7nfg8123. Review platform terms and local eligibility directly before using any trading service.
Evaluate OKX for your use case
Check regional eligibility, current fees and product availability on the official destination.
Review OKXAffiliate link · Availability varies by region · No guaranteed outcomeQuestions readers ask
How long can the memory boom last?
Based on the supplied brief, it can last while AI data center demand keeps exceeding available DRAM, HBM, and NAND supply. The brief does not prove a permanent boom, so the safer answer is that durability depends on price trends, new supply, and whether hyperscaler spending continues.
What is driving the current memory surge?
The supplied brief points to AI data center investment as the main demand source. It says GPUs, HBM, DRAM, and NAND-backed SSDs are being pulled into AI infrastructure, while memory makers prioritize high-margin data center products.
Is the boom mainly about higher shipment volume?
No. The brief says the biggest driver is the abnormal rise in memory prices. It describes DRAM spot prices and NAND wafer prices as rising roughly tenfold, which can sharply expand market value even without a matching rise in unit shipments.
Does this directly predict crypto prices?
No. The supplied brief does not name any crypto asset or on-chain metric. For crypto readers, the memory boom is better treated as AI infrastructure and macro technology context, not as a direct trading signal.
What should OKX users monitor next?
They should monitor DRAM and NAND pricing, hyperscaler capital expenditure, AI data center buildout pace, memory-maker capacity allocation, and signs of shortages in PCs and smartphones. Those checks are more useful than assuming the current growth rate will continue unchanged.
What is the biggest risk in the supplied thesis?
The biggest risk is cycle reversal. If supply catches up, AI infrastructure spending slows, or buyers resist higher component costs, memory prices could normalize and revenue growth could weaken quickly.