Starting today, the tool is accessible in the US and several other countries. This launch represents a significant milestone for Meta as it embarks on its journey to incorporate AI-driven video editing across its ecosystem, with more sophisticated features slated for release later this year.
Users can upload a 10-second video and select from over 50 AI-generated prompts — free of charge for a limited time. The tool can modify various aspects, including the user’s outfit, environment, mood, and overall scene aesthetic.
For instance, a rainy urban landscape can be transformed into a whimsical vignette featuring sparkles and soft-focus effects, or a selfie video can be reimagined as a vintage comic book panel or a high-energy video game scene.
Edited videos can be shared directly to Facebook, Instagram, or the Discover feed on the Meta AI app and website.
Meta emphasizes that this new feature is aimed at everyone, whether you are a casual user looking to enhance a family video or a content creator trying out a fresh visual style. The company collaborated closely with creators to develop prompt themes that resonate with online audiences.
“Our goal is to make high-quality video editing intuitive and accessible for all,” Meta expressed in the announcement. “We want to empower individuals to tell their stories and engage in creative experimentation.”
This launch builds on years of AI research at Meta. Following the 2022 release of the Make-A-Scene models for multimodal generation, the company introduced Llama-based diffusion models and Movie Gen — a system that allows for custom video generation and text-based editing.
The new editing feature, inspired by Movie Gen, is Meta’s inaugural consumer-facing product in the generative AI video realm. It sets the stage for a forthcoming update that will enable users to guide edits using custom text prompts.
Meta indicates that this update is anticipated to launch later this year.
Also Read: Meta taps top researchers from Google, Sesame for new AI lab
