Piniverse

Developer(s): Gabriel Buonomano (2023), Shiming De (2024),​ Alexander Kwan (2023), Yongcheng Liu (2023), Yihong Xu (2023)

This Piniverse showcase is a generative AI that takes user input in text form and generates a 3d environment from the description of objects and their spatial relationships to each other in the prompt. Currently, it renders these objects as point clouds. Alongside this showcase, we also constructed a prototype that allows users to interact with an AI agent to create “core documents”, such as concept documents and construction plans, and break them down into atomic tasks to be assigned to specialized AI workers. Lastly, we conducted a study into using generative AI such as these prototypes to enhance user workflows.