Anaflash and Samsung Collaborate on Developing Embedded Flash IP for Battery-Powered Applications

Anaflash, a leading embedded AI processor developer, has teamed up with Samsung foundries to create embedded flash IP that will be used for a range of applications. According to a report by eeNews, Anaflash has obtained exclusive rights to use single crystalline silicon embedded flash from the University of Minnesota as part of the development process.

The embedded flash IP technology was initially developed by Professor Chris Kim at the University of Minnesota, and is being developed for use in embedded non-volatile memory, counterfeit detection sensors, and neural-morphic computing cores. Anaflash is partnering with the university on memory computing AI and is looking to commercialize non-volatile memory for battery-powered applications such as medical wearables, wireless sensors, and robots.

Anaflash CEO and co-founder Seung-Hwan Song noted that flash can be built using standard logic flows without any process overhead, making it easy to deploy in "advanced process nodes" without increasing costs. To oversee the development of the technology, Anaflash has established a wholly-owned subsidiary called SemiBrain, which has signed a contract with Samsung foundries.

Recently, Anaflash completed a funding round that was led by Mirae Asset Venture Investment and We Ventures, though the funding amount was not disclosed. The Korea Advanced Institute of Science and Technology has already used Anaflash's logic-compatible embedded flash IP to manufacture an AI chip, which will be showcased at the Custom Integrated Circuits Conference in San Antonio, Texas from April 23 to 26.

This collaboration between Anaflash and Samsung foundries is expected to bring about exciting new developments in the field of embedded flash IP and memory computing AI, with potential applications in various industries, including medical wearables, wireless sensors, and robots.