Adobe’s ‘Commercially Safe’ AI Video Generator Powered By Firefly Is Here

Video screenshot via Adobe Video & Motion

Adobe has set its generative artificial intelligence prowess into motion with the release of its Firefly Video Model in limited public beta. The tech giant says this is the first generative AI model for video designed with commercial safety in mind, with a select group of users test-driving its features to provide feedback and help refine its capabilities.

Video via Adobe

The Firefly Video Model introduces several new features to streamline video creation. Generative Extend, available in the Premiere Pro beta, allows editors to lengthen clips, smooth transitions, and hold shots for precise timing, filling in awkward gaps. This addition aims to solve common editing challenges with AI assistance.

Video via Adobe

In the Firefly web app, Adobe is rolling out Text to Video and Image to Video capabilities. These features let creators generate video content from text prompts or animate still images. Users can adjust camera angles, motion, and zoom, offering greater control over their video output.

Adobe emphasizes that its Firefly models are trained on licensed content, including media from Adobe Stock and the public domain. Its AI technology also adheres to its AI Ethics guidelines of accountability, responsibility, and transparency.

“The usage of Firefly within our creative applications has seen massive adoption, and it’s been inspiring to see how the creative community has used it to push the boundaries of what’s possible,” expresses Ely Greenfield, chief technology officer of digital media at Adobe. “We’re thrilled to bring creative professionals even more tools for ideation and creation, all designed to be commercially safe.”

Release details for the Adobe Firefly Video Model

For now, the Firefly Video Model is in a free public beta, with Adobe set to announce more details on pricing and availability as the tool moves closer to a full launch.

Video via Adobe

Elsewhere

Besides the video-focused tools, Adobe has made improvements to its Firefly Image 3 Model, speeding up image generation by four times. Available now on the Firefly web app, this update ensures faster turnaround for creators working with still images. Meanwhile, the new Generative Workspace in Photoshop lets designers experiment with multiple concepts quickly, cutting down on the time it takes to bring creative ideas to life.

Adobe Illustrator users also get a boost with the Firefly Vector Model, which has seen enhancements like Generative Shape Fill and Text to Pattern. These new features offer more control over vector designs, allowing users to fine-tune their artwork by adjusting how tightly design elements are packed together.

On the enterprise side, Adobe introduced Firefly Services, a collection of APIs for businesses looking to scale creative production. This includes new tools in beta like AI-powered dubbing and lip-syncing, as well as Bulk Create, which automates large-scale editing tasks. These features aim to help companies manage high-volume workflows with greater efficiency.

Looking ahead, Adobe previewed Project Concept, a collaborative platform for real-time image remixing. For enterprises, the company is testing new Firefly Services, including AI-powered dubbing and lip-sync features, and bulk image editing tools.

[via Adobe]