According to sources, Microsoft is working behind the scenes on MAI-1: an LLM with more than 500 billion parameters. It would be the largest AI model Microsoft develops without interference from OpenAI.

The information believes it knows that Microsoft has big AI plans. An LLM called MAI-1 is being worked on behind the scenes. If The Information’s information is correct, the LLM would be officially announced at the Build conference on May 21st. With 500 billion parameters, the MAI-1 would be the largest model from Microsoft’s stable.

Microsoft poached almost the entire workforce of the start-up Inflection AI and also took over the intellectual property for $650 million. The model was completely developed from the ground up for indoor use. Project leader Mustafa Suleyman is co-founder of Google DeepMind.

Who has the biggest?

With a major LLM, Microsoft is taking a different approach. Previous models that the company developed itself were lightweight models with a “limited” number of parameters: Phi-3 Medium has fourteen billion parameters. The big advantage of small AI models is that they require less energy to operate. This reduces energy costs and makes mobile applications possible.

By working with OpenAI, there was no need for Microsoft to develop large models itself. The Copilot applications all run on GPT-4. It is not yet clear how Microsoft wants to integrate the new model into its own ecosystem. We may learn more during the build.

Microsoft has already poured billions of dollars into working with OpenAI and developing its own models. Despite these significant investments, the strategy does not appear to be harming Microsoft. Microsoft overtakes Apple as the most valuable company with a market value of more than three trillion dollars.