Stability AI has announced Stable Video Diffusion, an AI model that generates videos by animating existing images. The model is currently in a research preview and users must agree to certain terms of use. There are concerns about potential abuse of the model, as it lacks a built-in content filter. Stable Video Diffusion comes in two models, SVD and SVD-XT, which can generate videos at different frame rates. The models were trained on a dataset of millions of videos and then fine-tuned on a smaller set. However, the source of the training data is unclear, raising potential legal and ethical challenges. Despite limitations, the generated clips are of high quality. Stability AI plans to develop more models and a text-to-video tool for commercialization. The company recently raised $25 million in funding but is facing financial challenges and the departure of a key executive.
