Google Unveils New Specialized TPU Chips to Expand AI Hardware Portfolio
Google has announced two new variants of its eighth-generation Tensor Processing Units (TPUs), the TPU 8tx and TPU 8i Max, aimed at diversifying its in-house semiconductor capabilities. The TPU 8tx, developed in collaboration with AMD, focuses on confidential computing by integrating security directly into the hardware layer to protect sensitive data during AI training and inference. The TPU 8i Max is designed for high-memory workloads, doubling the capacity, processing power, and bandwidth of previous models to handle larger datasets during queries. This strategic move aims to reduce Google's reliance on external chipmakers like Nvidia as hyperscalers—including Microsoft and Amazon—increasingly develop proprietary silicon to support their growing AI infrastructure demands.
Tickers: $AMD $AMZN $GOOGL $MSFT $NVDA
Topics: Artificial Intelligence Cloud Computing Custom Silicon Hardware Launch TPU
Read the full article at TechCrunch
Part of the story Alphabet Expands AI Semiconductor Ambitions with New TPU Chips (6 articles)