Samsung Electronics has introduced a new generation of artificial intelligence (AI) memory technology that is expected to strengthen its position as the world's leader in the memory chip industry.
The announcement was made in conjunction with the Future of Memory and Storage (FMS) conference in Santa Clara, California, as demand for AI technology continues to grow rapidly.
The new technology introduced is known as V10 Bonding V-NAND (BV-NAND). The chip is developed with more than 400 layers and uses a new architecture known as wafer bonding.
This approach allows more data to be stored in a smaller space, thus increasing the storage capacity and performance of the chip compared to existing technologies.
The need for high-capacity memory is increasing as the use of AI is no longer focused on training large language models, but is also used to generate answers to user queries in real time.
NAND memory plays an important role in storing various data such as photos, videos and applications on smartphones and other electronic devices.
Samsung also introduced zHBM technology, which is a method of vertically stacking memory chips on top of AI processors.
This design shortens the distance data travels, allowing for faster processing, lower power consumption and reduced heat generated during operation.
According to Samsung, this wafer bonding technology has the potential to offer more than 10 times the memory density of conventional HBM5 memory.
In addition, it can improve energy efficiency by up to three times and reduce thermal resistance by more than 50%, thus helping AI data centers operate more efficiently.
Despite concerns that demand for memory chips may slow if smartphone sales in China continue to decline, analysts believe that AI growth will continue to be a key driver of the industry.
Samsung also said that long-term agreements with customers could potentially contribute between 60% and 70% of the company's total memory sales in the future.
At the same time, Citi analysts said that the supply of DRAM and NAND chips is still at a low level, thus indicating that demand for AI technology remains strong.
