Samsung and Arm partner on 2nm chips for on-device AI
Samsung Electronics and Arm have entered a partnership to develop next-generation on-device AI SoCs using 2nm process nodes. The collaboration aims to shift generative AI processing from cloud data centers to edge devices like smartphones to reduce latency and infrastructure reliance.
Key Takeaways
- Arm provides AI accelerator architecture and RTL design while Samsung manages SoC integration and manufacturing.
- Production will leverage Samsung's advanced 2nm process nodes for mass-market commercialization.
- The collaboration uses a custom chip model where finalized hardware is delivered directly to end customers.
- Arm shares rose approximately 2.5% in U.S. pre-market trading following the announcement.
Why It Matters
This partnership signals a strategic shift toward localizing generative AI workloads, reducing the streaming and mobile industries' heavy reliance on expensive cloud infrastructure. By moving computation to the device level, Samsung and Arm aim to eliminate data transmission latency that currently hampers real-time AI video and image applications. For the broader ecosystem, this move pressures competitors to accelerate their own 2nm roadmaps to maintain performance parity in the high-end smartphone market. Watch for the first production yields from Samsung's 2nm foundry lines to determine if this hardware can meet the power efficiency demands of sustained on-device generative processing.
Additional Context
The industry is increasingly prioritizing local AI workloads to bypass the bandwidth constraints of traditional cloud-based rendering and processing.
Read full article at tradingkey.com
Enjoy our coverage?
Add StreamingMeme as a preferred source on Google to see more of our streaming news at the top of your Search results.
Add as preferred source