3 articles tagged with #power-efficiency. AI-curated summaries with sentiment analysis and key takeaways from 50+ sources.
Normal Computing has raised $50 million in funding from Samsung Catalyst to develop new computing architectures that address the high costs and power consumption of current AI chips. The startup is part of a growing movement of companies seeking alternatives to GPU-based AI computing through innovative architectural approaches.
Major AI data center operators like Microsoft are turning to high-temperature superconductors (HTS) to address power efficiency challenges, as AI infrastructure demands exceed grid capacity. Microsoft invested in Veir's $75M Series B funding round to develop superconducting power cables that can deliver 10x higher capacity than conventional lines while eliminating transmission losses.
Researchers have developed TRAM, a technique that jointly optimizes low-power approximate multiplier structures with AI model training parameters, achieving up to 27% power reduction in vision transformers without significant accuracy loss. This approach differs from prior methods by integrating hardware design with model training rather than designing multipliers separately.
