Microsoft is reportedly creating an internal AI language model large enough to compete with those from Alphabet’s Google and OpenAI, according to a story in the Information.
The recently hired co-founder of Google DeepMind and former CEO of the AI company Inflection, Mustafa Suleyman, is reportedly in charge of the new model, internally known as MAI-1, according to the story, which quoted two Microsoft employees with knowledge of the project.
The model’s precise goal is yet unknown and will depend on how well it works. The source claims that Microsoft might introduce the new gadget later this month at its Build developer conference.
According to the report, MAI-1 will cost more since it will be “far larger” than the earlier, smaller, open-source models that Microsoft had trained beforehand.
Microsoft released Phi-3-mini, a scaled-down artificial intelligence model, last month in an effort to reach a larger customer base with more affordable choices.
The business has poured billions of dollars into OpenAI and integrated the technology of ChatGPT’s creator into its whole productivity software package, giving it an early advantage in the fight for generative AI.
The article claims that Microsoft has been allocating a sizable cluster of computers outfitted with Nvidia graphics processing units in addition to copious amounts of data to enhance the model.
In accordance with the analysis, MAI-1 should have about 500 billion parameters, GPT-4 from OpenAI should have one trillion, and Phi-3 small should have 3.8 billion parameters.
Microsoft added additional Inflection employees and named Suleyman as the head of its newly formed consumer AI branch in March.
Additionally, the study noted that although the new model might use initial training data, it does not use data from Inflection.
