Urgent.News

What's breaking now, across thousands of outlets.

AI

Memory price skyrockets as demand for AI data centers drives 369% year-over-year surge

Memory makers aren't selling much more; they're charging much more.

Memory price skyrockets as demand for AI data centers drives 369% year-over-year surge

Global DRAM revenue jumped 369% year over year in Q3 2026 to nearly $200 billion, driven by soaring demand for AI data centers, according to Counterpoint Research. The surge in prices, rather than volume, is the primary cause, as shipments increased only about 4%. Samsung, SK hynix, and Micron reported record profits, while phone and PC shipments declined around 20%.

The price of a single 16Gb DDR5 chip rose from $8.70 in October 2025 to $48 in September 2026, representing an increase of over 450%. HBM4 prices are also projected to reach $4 per gigabit. The memory industry is transitioning from a commodity to a scarce resource as AI demand escalates, causing memory makers to charge a significant premium over previous pricing.

Written by urgent.news from TechRadar's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

Read the original at techradar.com →

More in AI

Building Autonomous AI Research Agents with LangGraph, ReAct & Memory

Simple LLM chains ( Chain.run() ) are linear and brittle: if a tool call fails or a web search returns an empty snippet, the entire pipeline halts.

  • LangGraph enables cyclic state machines with state persistence for autonomous AI research agents.
  • ReAct pattern combines Reason, Act, and Observation steps with hierarchical supervisors and critics.

How to Build Enterprise AI Applications with Hybrid RAG & Production Evals

Over 80% of enterprise generative AI projects stall in prototype phase. Why? Because naive vector search with cosine similarity works great in a Jupyter notebook with 50 documents, but completely…

  • Over 80% of enterprise AI projects fail in prototype phase
  • Hybrid RAG combines sparse and dense retrieval methods
  • Production evaluation metrics: Context Recall, Faithfulness, Answer Relevance

When We Can No Longer Understand AI Decisions: Why the Future Needs Blockchain

We tend to think of AI development as a race toward greater intelligence: models are becoming smarter, faster, and capable of processing ever-increasing amounts of information.

  • AI decisions may become incomprehensible to humans as AI grows in complexity.
  • Blockchain can provide data integrity and security for AI decision-making.
  • Integrating AI with blockchain ensures accountability and transparency in AI systems.

More from Wednesday 11 November →