Kevin Scott Discusses Microsoft's AI Strategy on Podcast

Kevin Scott, Chief Technology Officer at Microsoft, detailed the rationale behind the tech giant's partnership with OpenAI, established in 2019.

by Faruk Imamovic
SHARE
Kevin Scott Discusses Microsoft's AI Strategy on Podcast
© Getty Images/Jerod Harris

In a revealing episode of Reid Hoffman's podcast "Possible," Kevin Scott, Chief Technology Officer at Microsoft, detailed the rationale behind the tech giant's partnership with OpenAI, established in 2019. Describing the alliance as "basically a bet," Scott emphasized the strategic decision to back a team known for its disciplined approach to scaling computing power and achieving remarkable feats with technology.

"This partnership with OpenAI was essentially a bet on this particular team, which at the time also understood that this was a game of scaling, compute, and doing incredible things with it in a very disciplined way," Scott explained.

Navigating Competitive Waters

Since taking on the role of CTO in 2017, Scott recognized the rapid acceleration in artificial intelligence (AI) development. To keep Microsoft competitive, he saw the urgent need to enhance the company's infrastructure.

This insight led to strategic planning that began as early as 2018, culminating in a pivotal email to Microsoft cofounder Bill Gates and CEO Satya Nadella the following year. Scott expressed deep concerns about Google's advancing AI capabilities in his email, which was made public during the Department of Justice's antitrust case against Google.

The email, titled "Thoughts on OpenAI," was sent just weeks before Microsoft declared its significant $1 billion investment in OpenAI. The decision to partner with OpenAI was also driven by the recognition that Microsoft could not tackle every challenge independently.

"And I was like, 'if we work with them, they will push us to build better infrastructure, and we can enable them to do their best work'," Scott remarked on the podcast.

Microsoft's Ongoing Innovations

Further affirming Microsoft's commitment to AI, a recent report by The Information revealed that the company is developing its own large language model (LLM) named MAI-1, which will boast roughly 500 billion parameters.

This development is part of Microsoft's broader strategy to compete with industry giants like Google and even OpenAI itself. Scott confirmed the development of MAI-1 in a LinkedIn post, indicating that Microsoft occasionally needs to conduct its own custom work, whether that involves training a model from scratch or fine-tuning an existing one.

"There will be more of this in the future too. Some of these models have names like Turing, and MAI. Some, like Phi for instance, we even open source," he added.

Microsoft
SHARE