Google Unveils Two New AI Chips to Challenge Nvidia Dominance

Google announced two new artificial intelligence chips, TPU 8tx and TPU 8i Max, to strengthen its position against Nvidia in the AI processor market. The announcement was made by Amin Vahdat, Senior Vice President and Chief Technology Officer responsible for AI and Infrastructure.

TPU 8tx is designed for processing sensitive data without moving it between different systems. The chip keeps data protected during processing, not only while stored or transmitted, by integrating security mechanisms directly at the hardware level.

“Google aims to challenge Nvidia's leadership in the AI chip market with these moves.”

According to Google, TPU 8tx can reduce latency by up to two times compared to TPU 8t without significant performance losses from additional security mechanisms. The system combines Google TPU technology with fifth-generation AMD EPYC processors and uses an optimized communication infrastructure.

TPU 8i Max was developed for larger and more complex workloads. It supports continuous operation of complex AI agents, analysis of extensive software codebases, and fast processing of long video data thanks to high memory capacity and processing power.

Google has introduced four AI processor systems within about one year. The company states these moves aim to challenge Nvidia's leadership in the AI chip market while offering businesses more flexible, secure, and cost-effective alternatives for training and running AI models.

Alphabet shares, Google's parent company, rose by more than 3 percent in the first hours of trading after the processor presentation. The announcement came on Monday, November 9, 2026, during a period of rapidly growing demand for AI infrastructure at Google Cloud.