Adobe announced on Monday that it has publicly released an AI model capable of generating video from text prompts. This move puts Adobe in competition with OpenAI’s Sora, which was introduced earlier this year, as well as video tools from TikTok owner ByteDance and Meta Platforms.
The AI model, known as the Firefly Video Model, is part of Adobe’s strategy to develop models trained on data that it has the rights to use, ensuring that the output can be legally used in commercial work. While Adobe has not yet specified a general release date, it has started offering the tool to people on its waiting list.
Although Adobe has not disclosed any customers using its video tools, it revealed that Gatorade, owned by PepsiCo, will use its image generation model for a site where customers can order custom-made bottles. Additionally, Mattel has been utilizing Adobe tools to help design packaging for its Barbie line of dolls.
Ely Greenfield, Adobe’s chief technology officer for digital media, explained that the focus of their video tools is to make them practical for everyday use by video creators and editors. The company has placed a special emphasis on ensuring that the generated footage seamlessly blends in with conventional footage by providing fine-grained control and teaching the model concepts used by video editors and videographers, such as camera position, angle, and motion.