Researchers at MIT have developed a new AI system capable of generating highly realistic and controllable 3D models from simple text descriptions. The system, named ‘DreamCraft3D’, uses a two-stage process that first creates a coarse 3D model and then refines it with detailed textures and geometry. This approach allows for greater artistic control and higher fidelity outputs compared to previous methods, which often produced blurry or inconsistent results. The technology has potential applications in video game development, film production, and virtual reality, enabling faster creation of digital assets. The team has released a paper detailing their work and plans to make the code publicly available for further research and development. Read the full article at https://technologyreview.com/2024/03/15/dreamcraft3d-ai-3d-modeling.