Microsoft president and vice chair Brad Smith (pictured) outlined a series of principles for the responsible deployment of AI, as well as a multi-year partnership with France-based Mistral AI during a keynote address today at MWC24 Barcelona.

A representative from Microsoft told Mobile World Live the company made a $16 million investment in Mistral AI, which would convert into equity in Mistral’s next funding round.

The collaboration with Mistral AI is notable due to Microsoft’s previous investments in rival OpenAI, which is believed to total roughly $13 billion.

Smith stated the partnership with Mistral AI reflects Microsoft’s commitment to work with a broad range of companies as part of its AI Access Principles announced at MWC24.

Mistral AI is now able to train and deploy its AI models across Microsoft’s data centres. Its AI models will be available to developers, or anyone else that wants to use them, through Microsoft’s Azure cloud service via its “as a service” initiative.

“We’re focused not just on proprietary software, not just on our partnership with OpenAI, as critical as it is, but with a wide number of companies,” Smith stated. “It’s not just training and deploying models. It’s about helping people bring their models to customers, to software developers so they can use them and the customers so they can buy them”.

Smith noted Microsoft has almost 1,600 AI models running across its data centres, with 1,500 of them open-source. He stated anyone that wants to create a model, or create an application that runs on a model, has access to the public versions of the APIs Microsoft uses.

The company didn’t provide the financial details of its partnership with Mistral AI.

According to a blog by Smith, Microsoft’s AI principles govern how it operates its data centres and additional AI assets located around the world. While there are a total of 11 AI principles, he explained they fall within three primary categories.

The first is the responsibility to enable innovation and competition while the second is meeting AI obligations under the laws and regulations in each country. The last principle is creating a broad array of AI-based partnerships, which includes the joint development of technology stacks.

“They’re about access. They’re about fairness and they’re about our broader societal responsibilities,” he stated.

He stated customers are also free to take their models and data to other companies’ data centres or clouds.

“It’s our responsibility to enable them to easily do so.”