Microsoft Forms GenAI Team to Focus on Building Smaller, Cheaper AI Models
-
According to The Information, Microsoft is forming a team called GenAI with the goal of developing smaller and more affordable AI models. These models will focus on language models, similar to large language models (LLMs) like OpenAI's GPT-4, but will require fewer computational resources.
This initiative aims to save Microsoft hundreds of millions of dollars in expenses required to run AI models on supercomputers. The GenAI team will be led by Microsoft Corporate Vice President Misha Bilenko and will become part of the Azure cloud division. Microsoft aims to improve the scalability and economic viability of AI models by developing smaller, more efficient language models.
This represents a significant development for companies providing AI services in the cloud, as smaller models can run on a wider range of hardware with lower costs, making them easier to deploy and apply.
