The ongoing memory crisis continues to leave a clear mark on consumer technologies. In particular, the increase in memory costs in computer hardware, combined with the demands of AI-focused data centers, is increasing pressure on the sector. While this demand increase is not expected to decrease in the near future, production capacity is now supporting the automotive sector as well. Micron's management team states that the RAM demand for autonomous vehicles could grow on a terabyte scale instead of kilograms. The company's latest financial results also show that AI-focused demands are transforming the sector: Micron achieved revenue of $23,86 billion in the second quarter of the year, registering approximately 200% growth compared to the same period of the previous year. Among the main drivers is the strong demand for high-bandwidth HBM memory chips. In response to the increasing demand, Micron aims to expand its capacity with large-scale production facilities planned in Japan, Singapore, and New York in the US. In addition to investments to be implemented in 2028-2029, they aim to increase production capacity by 20% by 2026.
Memory requirements for autonomous driving are skyrocketing. Autonomy capabilities in the automotive industry span a wide spectrum from L0 to L5. While current L2 systems partially automate steering and cruise control with driver-assisted technologies, L4 systems rely on autonomous driving capabilities, performing basic driving functions without a driver. At this point, the use of high-capacity and fast RAM for data processing is vital; vehicles essentially become supercomputers. Players like NVIDIA are advancing autonomous driving solutions with BYD, Geely, Isuzu, and Nissan through their Drive Hyperion platform. It's now clear that high-speed and large memory capacity plays a key role in this ecosystem, and Micron's 300 GB RAM target makes sense in this context. Although current car models generally have at least 16 GB of memory, this figure is expected to increase exponentially as autonomous driving becomes more widespread. When manufacturers begin producing hundreds of thousands or even millions of L4 autonomous vehicles, a significant increase in global memory demand and a potential supply crisis may arise. While we are still in the early stages of the L4 journey, technological advancements continue unabated. SK Group Chairman Chey Tae-won states that supply chain problems may persist for the next 4-5 years and that the industry will reach a more stable position by 2030.
