Nerds rejoice: Google just released its internal tool to collaborate on AI

Google is making machine learning frictionless.
Google is making machine learning frictionless.
Image: Google
We may earn a commission from links on this page.

As if giving the world its AI framework wasn’t enough, Google is now letting others work with a once-internal development tool, Colaboratory.

The software works a lot like Google Docs, its document collaboration tool, but with the ability to run code and show that code’s output within the document. Colaboratory is free and built on top of the open-source Jupyter project, software often used in data science.

It’s easy to take a skeptical view of anything free—Google stands to benefit if its software becomes a standard for teaching students to work in machine learning and data science. A similar strategy can be seen with TensorFlow, the AI framework Google open-sourced, which made it so easy to get started running AI software that a cucumber farmer built his own sorting system with it. Of course, that also means another customer for its Google Cloud APIs, per-use AI software sold by the company.

Google is also working to teach other companies to use its AI services; Colaboratory is pitched as an education tool as well as collaboration for research.

With Colaboratory, users create notebooks, or documents, that can be simultaneously edited like Google Docs. It supports Python 2.7 and has to be used on Google Chrome. The software is also integrated with Google Drive, so users can easily share projects or copy others’ shared projects onto their own accounts.

You can tinker with Colaboratory now, but Google is initially limiting the number of notebooks that can be created, and users must apply to be connected to Google’s backend to run code within a notebook.