SoftBank’s Saimemory joins forces with Intel on next-gen memory

2026-02-04 · Source: Tech Monitor · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Emerging Technologies & Innovation · Depth: Intermediate, quick

Summary

SoftBank's subsidiary, Saimemory, has partnered with Intel to commercialize Z-Angle Memory (ZAM), a next-generation memory technology. This collaboration aims to meet the increasing demand for computational performance in AI and high-performance computing by improving memory capacity and bandwidth while reducing power consumption. The alliance targets operational commencement in early 2026, prototype development by fiscal year 2027, and full commercialization by fiscal year 2029. Saimemory, established in December 2024, will lead ZAM's commercialization, while Intel will provide technology and standards collaboration. ZAM technology builds on the US Department of Energy's Advanced Memory Technology (AMT) research program and Intel's Next Generation DRAM Bonding (NGDB) initiative, which demonstrated increased DRAM density, bandwidth, and reduced latency and energy consumption.

Key takeaway

For AI Hardware Engineers designing future systems, this collaboration signals a significant shift in memory architecture. You should evaluate Z-Angle Memory (ZAM) as a potential solution to overcome current AI system scaling bottlenecks, particularly concerning memory capacity, bandwidth, and power efficiency. Consider how ZAM's planned commercialization by fiscal year 2029 could impact your long-term hardware roadmaps and design choices.

Key insights

Intel and SoftBank's Saimemory are commercializing Z-Angle Memory to address AI and HPC memory bottlenecks.

Principles

• Next-gen memory requires novel architectures.
• Collaboration accelerates technology deployment.

Method

The ZAM commercialization plan involves operational commencement (early 2026), prototype development (FY2027), and full commercialization (FY2029), leveraging prior AMT and NGDB research.

In practice

• Integrate ZAM for AI system scaling.
• Explore stacked DRAM concepts for performance.

Topics

• Z-Angle Memory
• Next-Gen Memory
• AI Computing
• Advanced Semiconductor Technology
• DRAM Architecture

Best for: AI Hardware Engineer, Executive, Research Scientist

Related on AIssential

• This chip startup just raised $135M on a bet that AI’s biggest bottleneck isn’t compute — it’s memory — AI inference is increasingly a memory scaling problem, not solely a compute challenge.
• China Just Built What TSMC Said Was Impossible — The next leap in computing performance may come from optimizing data movement and 3D packaging, not just transistor shrinkage.
• Building an Offline “Life Memorizer” with Gemini 2.0 & Qdrant Edge — On-device multimodal memory systems can be built privately using unified embeddings and embedded vector databases.
• The Memory Wall Is Real, Here Is the Door — Persistent DRAM shortages and rising costs necessitate hardware memory compression for AI product roadmaps.
• Abishai Financial Asia Eyes AI Memory Supercycle - Analytics Insight — Agentic AI architectures are driving unprecedented demand for high-bandwidth memory, creating a "supercycle" in semiconductor equities.
• NAND Reimagined in High-Bandwidth Flash to Complement HBM — HBF stacks NAND dies for high-capacity, high-throughput memory, complementing HBM in AI inference architectures.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Tech Monitor.