Google will allow you to put AI chips in your computer from October

Google, one of the leading companies in the field of artificial intelligence, will begin to allow its users to use their custom processors for this technology from October.

Google TPU, or tensor processing units, speed up artificial intelligence tasks such as understanding voice commands or recognizing objects in photos. Today, Google allows you to pay for these tasks to perform on your cloud computing infrastructure. But, thanks to a program called Edge TPU that was announced Wednesday, Google will allow programmers to install TPUs on their own machines.

"There are many benefits to making intelligent decisions in real time, to the point where these devices connect to a network", without having to wait for them to visit the Google machine network, said Injong Rhee of the Google Cloud division in a blog. "Your sensors will become more than data collectors, but they will make intelligent decisions in real time."

Artificial intelligence, often called machine learning or deep learning, represents a major change in computer science. It allows users to train computers to use real-world data to determine models by themselves.

Google offers artificial intelligence services in its cloud computing infrastructure, which allows customers to pay using technology. Allowing customers to use Google's artificial intelligence chips on their own computers could dramatically increase the number of users and the number of jobs by focusing on technology Google AI.

This strategy also places Google as a serious opponent of the Microsoft Azure service.

Google TPU will be available in a hardware module that can be connected to a computer using a PCI Express expansion slot (common on servers) or a USB port. Interested Google Cloud customers can apply online to access Google's AI chips.