Researchers at Cornell Tech have developed a novel optical receiver that can directly alter a processor's memory using light, potentially reducing the energy demands of AI systems. This technology bypasses traditional analog circuits used in optical receivers by accepting digital, QR code-like light matrices. The system aims to address the bottleneck of moving data between memory and processors, which is crucial for scaling AI models in applications like data centers and robots. While currently a proof of concept, the team is working on increasing the speed and data transfer rates for real-world applications. AI
IMPACT Could reduce energy consumption for AI hardware, enabling more efficient AI deployment in edge devices and data centers.
RANK_REASON Novel technology presented at a research symposium. [lever_c_demoted from research: ic=1 ai=0.7]
- Cornell Tech
- Dennis Sylvester
- dynamic random-access memory
- IEEE/JSAP Symposium on VLSI Technology & Circuits
- Jae-Sun Seo
- New York City
- Sram
- University of Michigan
- Yifan He
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
