Get the latest tech news

Amidst the Noise and Haste, Google Has Successfully Pulled a SpaceX

In 2013 Google started work on TPUs and deployed them internally in 2015. Sundar first publicly announced their existence in 2016 at I/O, letting the world know that they’d developed custom ASICs for TensorFlow. They made TPUs accessible to outside devs via Google Cloud in 2017 and also released the second generation that same year. […]

OpenAI got a lot of attention with GPT4, a product based on the AIAYN paper, putting LLMs on the map globally, and Google has taken heat for not being the first mover. OpenAI last raised $6.6B at a $157B valuation late last year, which incidentally is the largest VC rounder ever, and they did this on the strength of GPT4 and a straight line trajectory that GPT5 will be ASI and/or AGI, or close enough that the hair splitters won’t matter. But as OpenAI is lining up Oliver Twist style asking NVidia if “please sir, may I have some more” GPU for my data center, Google has vertically integrated the entire stack from chips with their TPUs, to interlink, to the library (TensorFlow) to the applications that they’re so good at serving to a global audience at massive scale with super low latency, using water cooled data centers that they pioneered back in 2018 and which NVidia is getting started with.

Get the Android app