High-fidelity 3D geometries of the natural and built world around us are an essential part of answering some of the most pressing scientific questions of our day. Through advances in deep learning, computer vision, and artificial intelligence more broadly, much progress has been made in reconstructing real geometries from images and/or sparse data, but these methods are just beginning to be applied to scientific problems. On January 7th, 2024 we bring to-gether researchers from computer vision, applied mathe-matics, and several scientific disciplines, to discuss the state of the art in 3D geometry generation and how it can be applied to open problems in science. This paper summarizes a selection of the talks and papers that have been accepted.