Google accelerates TPU cadence for AI

Custom silicon pushes infrastructure beyond GPUs.

September 3, 2026

Jon Peddie

Google is accelerating its custom silicon program as AI infrastructure demand changes the economics of data center computing. The company moved from a two-year TPU cadence to two processors in one year, and AI infrastructure chief Amin Vahdat expects that pace to increase. Alphabet also raised 2026 capital spending guidance to $195–$205 billion. Google now faces a broader engineering challenge: coordinate processors, memory, networking, packaging, racks and suppliers quickly enough to turn silicon advances into deployable AI capacity. What the hell are these things? Google has spent more than a decade developing custom AI processors, starting with its first Tensor

...

Enjoy full access with a TechWatch subscription!

TechWatch is the front line of JPR information gathering service, comprising current stories of interest to the graphics industry spanning the core areas of graphics hardware and software, workstations, gaming, and design.

A subscription to TechWatch includes 4 hours of consulting time to be used over the course of the subscription.

Already a subscriber? Login below

Google accelerates TPU cadence for AI

Enjoy full access with a TechWatch subscription!

This content is restricted

Login