Alphabet's Google announced two new variations of its TPU processors November 9, 2026: 8tx for confidential computing in government and regulated industries, and 8i Max for data-heavy AI applications. The release adds competition to Nvidia's AI hardware business.
Google broadens its homegrown AI chip lineup
Google introduced the 8tx and 8i Max tensor processing units on Monday, November 9, as demand for cloud-based artificial-intelligence hardware continues to expand. Google has traditionally used TPUs within Search, YouTube and Gemini but is increasingly selling access to this technology through Google Cloud, challenging Nvidia's leadership in AI accelerators. Reuters reported that Alphabet shares gained more than 3% in early trading, a time-specific price move rather than an all-day result.
8tx targets privacy-sensitive institutions
The 8tx is built for government agencies and businesses in tightly regulated fields that need to protect information while computing. According to Google's claims reported by Reuters, it pairs Google TPU technology with AMD processors and hardware-level data encryption. Google says the redesigned system can deliver up to twice the performance of its predecessor with little performance penalty from security controls. These performance figures are manufacturer claims rather than independently verified tests across all workloads.
8i Max aims for unusually demanding AI models
Google says the 8i Max combines two chips into one package, offering more memory and compute for applications handling large quantities of context and data. The company cited coding assistants, mathematical reasoning and drug discovery as possible uses. Wider availability, customer adoption and real-world speed will depend on cloud access and workload requirements. CDN News will distinguish product announcements from deployment or claimed benchmark results.
Sources & transparency
Reuters — Google unveils new AI chip variants as cloud demand surges, November 9Original CDN News coverage. Sources are linked for reference.
Request a correction