Samsung Electronics has introduced its next-generation artificial intelligence (AI) memory technology, unveiling a prototype of its V10 Bonding V-NAND (BV-NAND) at the Future of Memory and Storage (FMS) Conference in Santa Clara, California.

The new memory technology is designed to strengthen Samsung's leadership in the rapidly expanding AI semiconductor market by delivering significantly higher storage density, faster performance, and improved energy efficiency for AI-driven applications.

400-Layer BV-NAND Boosts AI Storage Performance

Samsung's latest V10 BV-NAND features more than 400 memory layers and introduces a new wafer-bonding architecture that allows memory cells to be stacked more efficiently.

According to the company, the technology delivers:

• Around 58% higher memory density than the previous V9 NAND generation.
• Faster read and write speeds.
• Improved input/output performance.
• Better power efficiency for AI workloads.

The innovation is aimed at supporting growing AI applications that require massive amounts of high-speed data storage, including generative AI, large language models, cloud computing, and enterprise data centers.

AI Demand Driving Memory Innovation

As artificial intelligence evolves beyond model training to real-time inference and AI-powered services, demand for high-capacity NAND flash memory continues to rise.

NAND memory stores digital data such as applications, photos, videos, and AI datasets across smartphones, PCs, servers, and cloud infrastructure.

Samsung said its latest technology has been specifically developed to meet the increasing storage and performance requirements of AI systems.

Samsung Introduces zHBM and zNAND-O Concepts

In addition to BV-NAND, Samsung showcased concept designs for what it described as the industry's first zHBM and zNAND-O architectures.

Unlike conventional High Bandwidth Memory (HBM), which sits beside AI processors, Samsung's zHBM places memory directly above AI accelerators through vertical stacking.

The company says this design offers:

• Higher memory bandwidth.
• Lower latency.
• Improved power efficiency.
• Reduced heat generation.
• Shorter data transmission paths.

Samsung estimates its wafer-bonding technology could eventually provide:

• More than 10 times the memory density of conventional HBM5.
• Up to three times greater energy efficiency.
• More than 50% lower thermal resistance.

AI Market Continues to Support Chip Demand

Despite concerns about slowing smartphone demand in some markets, analysts believe artificial intelligence will remain the primary driver of memory chip growth.

Samsung recently indicated that long-term customer agreements could eventually account for 60% to 70% of its memory business, highlighting strong confidence in sustained AI infrastructure investment.

Industry analysts also note that inventories across both DRAM and NAND supply chains remain below historical averages, supporting continued demand for advanced memory technologies.