Microsoft CEO Satya Nadella said the company’s partnership with OpenAI gives it access to all of OpenAI’s intellectual property except consumer hardware, providing Microsoft with AI capabilities that reduce concerns about Google’s competition in the field. OpenAI, on its part, has also struck deals with Nvidia, AMD and Google for AI chips, data centers and AI model training.During a podcast, Nadella explained the scope of Microsoft’s access to OpenAI’s technology. When asked if Microsoft has access to OpenAI’s IP, Nadella responded: “In our case, the good news here is OpenAI has a program in which we have access to.”When pressed about whether Microsoft has access to all of it, Nadella confirmed: “All of it.”“So the only IP you don’t have is consumer hardware?” the interviewer asked, to which Nadella responded, “That’s it”. “By the way, we gave them a bunch of IPs as well to bootstrap them, right? So, this is one of the reasons why they had a mass because we built all these supercomputers together, and they benefited from it. Rightfully so,” Nadella added.
Microsoft’s Maia chip delay
Microsoft has delayed mass production of its next-generation Maia AI chips until 2026, according to a previous report from The Information. The tech giant pushed mass production of the new AI chips, codenamed Braga, back by at least six months. Braga is expected to be renamed Maia 200 upon release.Microsoft had planned to deploy the Braga chip into data centers this year, but delays attributed to unanticipated design changes, staffing constraints and high turnover made this impossible.While Microsoft’s Maia chips face delays, Google’s Tensor Processing Units have become a preferred option for training AI models. The TPUs have established themselves as reliable hardware for AI development, giving Google an advantage in the AI infrastructure market.Google designed its TPUs specifically for machine learning workloads and has refined them through multiple generations. The chips are widely available through Google Cloud and have been used to train major AI models, making them a proven option for AI developers.
OpenAI’s chip strategy
In addition to influencing Microsoft’s chip development efforts, OpenAI is building its own AI chips in partnership with Broadcom. The AI company has also struck deals with Nvidia, AMD and Google for AI chips, data centers and AI model training.
About the Author
