Business · Companies · 22 hrs ago
Humanoid robot DRAM demand forecast to grow more than 20-fold by 2030
Humanoid robots use cameras and sensors to gather information and make decisions about how to move.
That work requires memory, including DRAM, which temporarily stores data for processing.
Counterpoint Research forecasts that total DRAM demand for humanoid robots will rise from about 8,300 terabits this year to about 172,000 terabits in 2030.
It also expects the average DRAM in each robot to increase from 19 gigabytes to 39 gigabytes over that period.
The forecast reflects both wider robot adoption and more memory-intensive AI models.
This could create a new market for memory makers such as South Korea’s Samsung Electronics and SK hynix, with demand also expected for low-power DRAM and high-capacity storage.
Counterpoint also expects more demand for computing chips, including Nvidia GPUs, as robot AI models advance.
Humanoid robot DRAM demand is forecast to rise more than 20-fold by 2030, according to Counterpoint Research.
Total demand is projected to increase from about 8,300 terabits this year to about 172,000 terabits in 2030.
The average DRAM capacity per humanoid robot is expected to grow from 19 gigabytes to 39 gigabytes over the same period.
The growth is attributed to wider robot adoption and rising memory needs as AI models become more advanced.
The forecast points to robots becoming a new market for memory chipmakers including Samsung Electronics and SK hynix.
- Who
- Counterpoint Research made the forecast. Memory chipmakers including Samsung Electronics and SK hynix could benefit.
- What
- Humanoid robot DRAM demand is forecast to grow more than 20-fold by 2030.
- When
- The forecast compares this year with 2030.
- Where
- Not stated
- Why
- Robot adoption is expected to expand, and each robot is expected to need more memory as AI models advance.
This story does not have two clearly opposing sides.
Humanoid robots will need a similar level of memory and storage
Micron CEO Sanjay Mehrotra said humanoid robots would need a level of memory and storage similar to Level 4 or higher autonomous vehicles.
Counterpoint Research estimates total humanoid robot DRAM demand at about 8,300 terabits and average DRAM capacity per robot at 19 gigabytes.
Counterpoint Research forecasts total humanoid robot DRAM demand of about 172,000 terabits and average capacity per robot of 39 gigabytes.
- Total DRAM demand this year
- About 8,300 terabits
- Total DRAM demand in 2030
- About 172,000 terabits
- Average DRAM per robot this year
- 19 gigabytes
- Average DRAM per robot in 2030
- 39 gigabytes
- Companies cited
- Samsung Electronics and SK hynix









