Foundation models refer to large AI models that are trained using vast datasets to perform various tasks. These models mark a departure from the smaller task-specific models that previously dominated the AI landscape. 

First popularized by the Stanford Institute for Human-Centered Artificial Intelligence, the term “foundation models” refers to how these large models serve as the foundation for developing more complex and refined models across diverse applications. For example, GPT serves as a foundation model, with successors like GPT-3 and GPT-4 being harnessed by hundreds of startups and established companies to develop task-specific models for applications ranging from content creation to marketing plan generation. 

This hub includes organizations that have built large AI models, whether for in-house product development, to drive AI research, or for commercialization via third-party API access and other means. Due to the prohibitive costs of training and running large AI models, the space remains strongly incumbent-driven and is led by Big Tech firms, followed by AI research labs.

Among the startups, OpenAI is a rare frontrunner in terms of commercializing foundation models. However, Big Tech firms like Google and Meta have largely refrained from releasing their models for public use and have instead leveraged these models to enhance the functionality of in-house products. Meanwhile, AI research labs like EleutherAI operate as non-profit organizations focused on open-sourcing AI development. 

Note: Additional sections (market sizing, detailed overview, incumbents, etc.) can be provided on request.