Microsoft Acquires Bonsai as Drive Into Artificial Intelligence Deepens

Microsoft (MSFT) is acquiring Bonsai, a Berkley, California-based artificial intelligence company that was in the portfolio of the information technology giant’s venture capital fund, M12.

Terms of the deal weren’t disclosed by the companies on Wednesday. In a blog post, Microsoft’s Gurdeep Pall, corporate vice president of business AI, said Bonsai’s “novel approach” allows subject matter experts — including those without aptitude in AI — to “specify and train autonomous systems to accomplish tasks.”

Pall used an example of when experts from Siemens earlier this year used Bonsai’s AI platform and machine teaching to autocalibrate a Computer Numerical Control machine, in which a computer manipulates machine-shop equipment, 30 times faster than the traditional approach.

“This represented a huge milestone in industrial AI, and the implications when considered across the broader sector are just staggering,” Pall wrote.

Microsoft has been investing more in artificial intelligence, developing tools and services through its Azure cloud computing system for AI and acquiring companies like Bonsai, GitHub and Montreal-based Maluuba in 2017. The deal for software development platform GitHub was announced earlier this month.

Bonsai’s platform along with simulation tools and reinforcement learning work happening in Microsoft’s research unit will all become “the simplest and richest AI toolchain for building any kind of autonomous system for control and calibration tasks,” Pall said.

The chain will run on the Azure cloud with its graphics processing unit and Brainwave, which the company described last month as “a hardware architecture designed to accelerate real-time AI calculations.”

“Models built with it will be deployed and managed in Azure IoT, giving Microsoft an end-to-end solution for building, operating and enhancing ‘brains’ for autonomous systems,” Pall said.

Leave a Reply

Your email address will not be published. Required fields are marked *