Microsoft Corp. (NASDAQ: MSFT) inches up in pre trading session Friday as the business is collaborating with Advanced Micro Devices Inc. on the chipmaker’s development into artificial intelligence processors as part of a multipronged campaign to get more of the highly prized components, according to news sources.
According to the people familiar with the subject, the businesses are collaborating to provide an alternative to Nvidia Corp., which leads the market for AI-capable chips known as graphics processing units. According to the sources, Microsoft is bolstering AMD’s efforts by giving technical resources and collaborating with the chipmaker on a native Microsoft processor for AI workloads codenamed Athena.
A Microsoft official, Frank Shaw, denied AMD is a component of Athena. “AMD is a great partner,” he says. “However, they are not involved with Athena.”
The agreement is part of a larger push to boost AI processing capacity, which is in high demand following the growth of chatbots like ChatGPT and other services based on the technology. Microsoft is a leading provider of cloud computing services as well as a driving force in the application of AI. The business has invested $10 billion in ChatGPT producer OpenAI and has promised to include similar functionalities throughout its whole software portfolio.
The move also underscores Microsoft’s growing participation in the semiconductor business. Under former Intel Corp. executive Rani Borkar, the business has been creating a silicon division for several years, and the group now employs over 1,000 people. Last month, The Information reported on Microsoft’s development of the Athena artificial-intelligence processor.
Several hundred of those individuals are working on the Athena project, and according to one of the insiders, Microsoft has spent around $2 billion on its chip efforts. However, the project does not portend a breakup with Nvidia. Microsoft expects to continue collaborating with that business, whose processors are the workhorses for training and running AI systems. It’s also looking for methods to obtain more Nvidia chips, emphasizing the critical scarcity that Microsoft and others are experiencing.
Microsoft’s partnership with OpenAI, as well as its own slate of newly released AI services, are needing more processing power than the firm anticipated when it acquired processors and built up data centers. Businesses are interested in using OpenAI’s ChatGPT service as part of their own products or corporate applications, and Microsoft has developed a chat-based version of Bing as well as new AI-enhanced Office features.
It is also modernizing previous tools, such as GitHub’s code generator. All of those AI programs run in Microsoft’s Azure cloud and necessitate the expensive and powerful processors provided by Nvidia.
This is also a high priority for AMD. “We are very excited about our opportunity in AI — this is our No. 1 strategic priority,” said CEO Lisa Su during the chipmaker’s earnings call on Tuesday. “We are in the very early stages of the AI computing era, with adoption and growth rates faster than any other technology in recent history.”
Su also stated that AMD has the capability of producing partially customized chips for its largest clients to utilize in their AI data centers.
Developing an alternative to Nvidia’s range will be a difficult undertaking. This firm provides a software and hardware bundle that works together, including processors, a programming language, networking equipment, and servers, allowing users to quickly enhance their capabilities.
That is one of the reasons Nvidia has grown so powerful. However, Microsoft is not the only company attempting to create in-house AI processors. Amazon Cloud purchased Annapurna Labs in 2016 and has since built two separate AI processors. Alphabet Inc.’s Google has its own training chip.
