Microsoft may be working on its own AI chip that can be used, for example, to train large language models. Insiders report this to The Information. The company would like to become less dependent on expensive chips from other suppliers.
According to anonymous sources from The Information, Microsoft has been working on its own AI chip, codenamed Athena, since 2019. A small group of Microsoft and OpenAI employees would already have access to the chip. The company hopes that Project Athena will outperform the chips it currently buys from other suppliers, according to The Information. That should save time and money on his AI work. Other large tech companies, such as Amazon and Google, are also working on their own AI chips.
According to sources, the chip is used to train large language models and support inference, for example for use in AI models such as GPT-4 that the company uses in Bing and Office. Microsoft may be able to make the chips more widely available ‘next year’. However, it is not known whether the chips will be available to customers using Microsoft Azure or whether they will be for internal use only. Microsoft would also have an internal roadmap with future generations.
Chips for training AI models are currently mainly produced by Nvidia. That company offers, among other things, its A100 and H100 accelerators. Microsoft would like to save costs on the purchase of AI chips with its own chips. However, the tech giant does not see its chips as a “broad replacement for Nvidia’s products,” according to the sources.
There have been reports for some time that Microsoft is working on its own chips. In 2020, Bloomberg stated that the company is developing Arm chips for use in servers and consumer Surface devices. However, those alleged chips have not yet appeared, although the company previously collaborated with Qualcomm on Arm chips for its Surface lineup.
