A new study from MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) demonstrates a significant advancement in AI’s ability to understand and reason about the physical world. Researchers developed a system that can infer the 3D geometry of objects from a single 2D image and then simulate how those objects would behave in various physical scenarios, such as being pushed or stacked. This represents a leap beyond previous AI models that could recognize objects but struggled with intuitive physics and spatial reasoning. The technology has potential applications in robotics, where machines need to interact with unfamiliar objects, and in augmented reality. The findings were published in the journal Science Robotics. For the full details, read the complete article at https://www.technologyreview.com/2023/10/26/1082345/ai-physics-3d-reasoning-mit/