The memory semiconductor market, which has grown around artificial intelligence (AI) data centers, is expanding its demand sources to include humanoid robots. As the number of robots that can perceive their surroundings and act autonomously increases, the memory capacity required for computations is also on the rise. Analysts predict that Samsung Electronics and SK Hynix will encounter new growth opportunities in the semiconductor market for robots, following their success in high-bandwidth memory (HBM).
According to market research firm Counterpoint Research, the demand for DRAM used in the 'brain' of humanoid robots is expected to rise from approximately 8,300 terabits (Tbit) in 2026 to about 172,000 Tbit by 2030, marking an increase of more than 20 times in just four years.
The memory capacity per robot is also expected to grow. The average DRAM capacity for humanoid robots is projected to more than double from 19GB this year to 39GB by 2030. This increase in memory capacity per unit, alongside rising robot shipments, is expected to accelerate overall market growth.
Advancements in AI technology are also driving up memory demand. Humanoid robots process information collected in real-time from cameras, LiDAR, and tactile sensors to determine their movements. The spread of end-to-end AI models that integrate various types of data and the increasing application of world models that learn and predict the surrounding environment are contributing to the growing memory requirements for data storage and processing.
In addition to memory, the market for computational semiconductors is also expected to expand. Counterpoint Research forecasts that as the AI models for robots become more sophisticated, the demand for processing chips, such as NVIDIA graphics processing units (GPUs), will increase. Custom semiconductors (ASICs) being developed by automakers like Tesla and Xiaopeng are also seen as growth factors in this market.
Power efficiency is crucial in the memory market for robots. Humanoid robots, which operate on batteries, need to process large amounts of data while minimizing power consumption. Consequently, low-power DRAM (LPDDR) and high-capacity NAND flash are expected to emerge as key demand sources. In the long term, three-dimensional (3D) DRAM technology is also anticipated to be a significant variable in the related market.
However, the actual growth of the memory market for robots will depend not only on demand forecasts but also on the commercialization speed of humanoid robots, production scale, and how quickly memory specifications for individual products improve.
Samsung Electronics and SK Hynix have the opportunity to extend the memory demand formed around AI data centers into the robotics industry. According to Counterpoint Research, in the second quarter of this year, SK Hynix held a 50% share of the global HBM market, while Samsung Electronics accounted for 33%. The ability to respond to the expanding demand for robot memory based on the technological competitiveness secured in the AI semiconductor market is expected to be crucial for diversifying future growth drivers.
* This article has been translated by AI.
Copyright ⓒ Aju Press All rights reserved.