Samsung Deploys AI to Design Mobile Processors

Samsung has begun using artificial intelligence to automate the process of developing computer microchips. It is one of the first semiconductor manufacturers to apply AI to chip design, writes Wired.

The South Korean corporation is using software from Synopsys, a leading producer of electronic design automation (EDA) systems for semiconductor manufacturers.

“What you see here is the first truly commercial processor designed by artificial intelligence,” said Aart de Geus, chairman and co-CEO of Synopsys.

A Samsung spokesperson confirmed that they are using Synopsys AI software to design their Exynos chips, on which Galaxy smartphones are built. The company did not confirm whether AI-designed chips have entered production, or in which products they might appear.

Industry analysts say the Synopsys tool will speed up chip development and enable new chip architectures. This could be made possible by years of work on advanced semiconductor devices that can be used as training data for AI algorithms, they say.

Earlier, Google and Nvidia had announced plans to deploy AI for processor design.

In August, Google unveiled a mobile processor with tensor cores for machine learning of its own design.

In June, TSMC announced an updated manufacturing process for the N5A chips, designed for running AI in autonomous vehicles.

In early June, Google revealed that reinforcement learning had been used to accelerate chip design from several months to six hours.

Subscribe to ForkLog News on Telegram: ForkLog AI — all the news from the world of AI!