Google Launches Specialized AI Chips to Reduce Reliance on Third-Party Hardware
· Technology · TechCrunch
Google has introduced two new variants of its tensor processing units, the TPU 8tx and TPU 8i Max, to expand its in-house AI hardware portfolio. The TPU 8tx, developed in collaboration with AMD, focuses on secure, confidential AI model training and inference by isolating sensitive data at the hardware level. The TPU 8i Max combines two existing inference-focused chips into a single package to double memory capacity, processing power, and bandwidth for large-scale workloads. These developments are part of Google's broader strategy to decrease dependency on external chip manufacturers for its cloud computing operations. The launch occurs as major tech firms and AI laboratories increasingly develop proprietary semiconductors to challenge the market dominance of Nvidia.
Why it matters
This move allows Google to offer more secure and efficient AI processing capabilities to its cloud customers, reducing the company's long-term reliance on Nvidia's hardware dominance.
Read the original report — TechCrunch
Join us on Telegram
Breaking news the moment it lands. At 10,000 members we ship the Android app.