On Monday, Adobe announced Firefly Video Model, a new AI-powered text-to-video generation tool that can create new videos from written prompts. It joins similar offerings from OpenAI, Runway, Google and Meta in an increasingly crowded field. Unlike the competition, Adobe claims that Firefly Video Model is trained exclusively on licensed content, potentially circumventing ethical and copyright issues that have plagued other generative AI tools.
Because of its licensed training data, Adobe calls Firefly Video Model “the first publicly available video model designed to be commercially safe.” However, the San Jose, California-based software company has not yet announced a general release date and during a beta testing period it is only granting access to people on a waitlist.
A sample video of Adobe's Firefly Video Model, provided by Adobe.
The new model has been in production since at least April 2023 and builds on techniques that Adobe has developed for its Firefly image synthesis models. Like the text-to-image generator, which the company later integrated into Photoshop, Adobe hopes to target Firefly Video Model at media professionals, such as creators and editors. The company claims its model can produce footage that blends seamlessly with traditionally created video content.
