NYT – Google Makes its Special AI Chips Available to Others


I actually thought access to the TPS chips had been made available at least a year ago, but perhaps that was only from a simulation perspective???

  • few years ago, Google created a new kind of computer chip to help power its giant artificial intelligence systems
  • Monday, the internet giant said it would allow other companies to buy access to those chips through its cloud-computing service
  • Google’s move highlights several sweeping changes in the way modern technology is built and operated
    • not just big internet companies. They are big hardware makers
  • As a way of cutting costs and improving the efficiency of the multibillion-dollar data centers that underpin its online empire, Google designs much of the hardware inside these massive facilities, from the computer servers to the networking gear that ties these machines together just as other internet companies are doing
  • In addition to its T.P.U. chips, which sit inside its data centers, the company has designed an A.I. chip for its smartphones
  • Google’s new service is focused on a way to teach computers to recognize objects
  • to accelerate its work on driverless cars, Lyft began testing Google’s new chips to identify street signs or pedestrians. “Training” these systems can take days, but with the new chips, the hope is that this will be reduced to hours
  • T.P.U. chips have helped accelerate the development of the Google Assistant, the service that recognizes voice commands on Android phones, to Google Translate
  • also reducing Google’s dependence on chip makers like Nvidia and Intel.
  • It also designed its own servers and networking hardware, reducing its dependence on hardware makers like Dell, HP and Cisco
    • the only way to build an efficient service is to build your own hardware (interesting considering so much focus the last 10 years has been on software)
  • new wave of artificial intelligence driven by “neural networkscomplex algorithms that can learn tasks on their own by analyzing vast amounts of data
    • Typically, engineers train these algorithms using graphics processing units, or G.P.U.s (usually supplied by companies like Nvidia, Intel, etc)
    • “Google has become so big, it makes sense to invest in chips,”

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s