Draft poster · the finished cut will appear here
What happened
Adobe says Premiere’s Generative Media Tool can create video and sound effects directly where an editor needs them in the timeline. The company says the generated clips remain editable, and that Premiere can use Firefly or partner models without sending the editor to a separate application.
How it works
Adobe describes a context-aware workflow: a reference frame from the project can guide footage that fills a gap, while voice-guided sound effects, Generate Soundscape, and Generate Music add audio options in the same edit. Generate Soundscape and Generate Music are listed as beta features. Adobe also says After Effects has a public-beta AI Assistant that can inspect a project, reorganize large layer sets, troubleshoot problems, and create expressions.
The important change is where the model sees context. Instead of receiving a blank prompt and returning a detached asset, the tools are positioned beside the surrounding cuts, layers, timing, and audio. Adobe says the editor still chooses what to hand off, refine, and apply.
Why it matters
Generation becomes more useful when it participates in the artifact that already contains the creative decisions. A clip can be judged against neighboring shots, and an assistant can act on the structure that makes a composition work. That does not remove editorial judgment; it gives the model a more specific surface to operate on.
What we don’t know yet
Several capabilities are beta or public beta. Availability, partner-model terms, and model behavior may vary, and Adobe’s commercial-safety descriptions are product claims. This is not an announcement of an unsupervised editor.
Source: Adobe Blog, Deepa Subramaniam, September 8, 2026 - original post