Adobe Premiere Pro now includes AI tools that can generate new video frames at the edges of a clip — for example, extending a shot by a second or two so it fits a timeline gap — as well as AI-powered text-based editing, automatic captions in 27+ languages, and noise removal from speech recordings. The Generative Extend feature launched in beta in October 2024 and reached general availability in April 2025.

Generative Extend is powered by the Firefly Video Model, which was trained on licensed footage, and can add up to approximately two seconds of new frames at clip boundaries with matching motion and ambient audio. Text-Based Editing converts speech to a searchable transcript, letting editors cut footage by deleting words from the transcript. Media Intelligence (beta, 2025) enables semantic search — finding footage by describing what is happening visually, not just by file name. AI-generated frames carry Content Credentials.