Updated

Google unveiled two new tensor processing units, TPU 8tx and TPU 8i Max, extending its in-house AI chips to confidential training and memory-heavy workloads. TechCrunch reported the launch on November 10, 2026, describing it as part of Google's effort to reduce its dependence on chips made by other companies for its cloud business. The two chips are variants of the eighth generation introduced in April. TPU 8tx, designed with AMD and based on the TPU 8t, is aimed at training and running AI models confidentially, with security built into the hardware to keep sensitive data and model weights isolated during use. Its custom inferencing network lets it run reasoning alongside training, which Google said could help distill and train small models on the fly without moving proprietary data between devices. TPU 8i Max combines two of the inference-focused TPU 8i chips launched in April into one package, doubling memory capacity, processing power and bandwidth so models can handle a larger range of information during queries. TechCrunch linked the launch to hyperscalers and AI labs building their own chips to reduce reliance on Nvidia, and said the news came at a vulnerable moment for Nvidia after CEO Jensen Huang admitted over the weekend to an affair with someone alleged to be a Chinese spy. The article gives no pricing or availability dates.