Adobe is stepping up its AI game by adding video generation to its Firefly model, the company revealed this Wednesday. Up until now, Firefly has been known for its image creation and editing capabilities, which are integrated across Adobe’s Creative Cloud programs. But now, Adobe is expanding those capabilities to include video content.
In a teaser video, Adobe gave a glimpse of what’s coming. The video shows someone entering a detailed prompt into a text box, which then generates different kinds of videos. There’s also a sidebar where users can tweak the motion of a scene or even add reference images to the prompt for more customization. One of the exciting features shown is the ability to create new video footage to extend existing clips and seamlessly fill any gaps.
That’s all the information available for now, as this feature won’t be ready for release until later this year. If you’re eager to try it out, you can join the waitlist, but remember to sign up using the email tied to your Adobe account.
Earlier this summer, Adobe rolled out upgrades to its AI tools in Photoshop and Illustrator. These updates included an AI image generator for Photoshop and a tool for generating shapes in Illustrator. Adobe has made it clear that Firefly’s models are trained on licensed datasets, such as Adobe Stock and public domain images, but they don’t use personal content from Creative Cloud subscribers for training.
Adobe has faced some bumps along the way with its generative AI rollout. Earlier this year, some users were worried after an unclear update to Adobe’s terms of service led to fears that the company might be able to scan or access their content.
Adobe later clarified that it only moderates content stored in the Cloud and not on users’ devices. This content moderation is part of the effort to prevent illegal use of its AI tools. On top of that, Adobe is dealing with a lawsuit from the U.S. Department of Justice, which claims the company’s cancellation process is difficult and includes hidden fees.
