Meta for adding new AI-powered video features to apps

Meta has unveiled generative artificial intelligence (AI) research that shows how simple text input can be used to create custom videos and sounds and edit existing videos.

This AI model, called “Meta Movie Gen,” builds on the company's previous generative AI models Make-A-Scene and Llama Image, the company said in a blog post Friday (October 4).

Movie Gen combines the modalities of these previous generation models and, according to the post, allows even more detailed control.

Meta is currently making this new model available to only some employees and external partners, but plans to integrate it into some of its existing apps next year, Bloomberg reported Friday, citing an interview with Connor Hayes, a Meta vice president specializing in generative AI products.

Before Movie Gen is released to the public, Meta wants to prevent the technology from being used to create videos of people without their consent, according to the report.

Meta said in a blog post in April that three of its platforms would label AI-manipulated content as “Made with AI.”

In Meta's blog post on Friday, the company said: “While there are many exciting use cases for these base models, it is important to note that generative AI is not a replacement for the work of artists and animators.” We share this research because we believe in the “We believe in the power of this technology to help people express themselves in new ways and give people opportunities they might not otherwise have.”

Movie Gen's video generation capabilities allow it to create 16-second videos at 16 frames per second, taking into account object motion, subject-object interactions and camera movement, the post says.

The foundation model can also generate personalized videos based on a person's image and a text prompt, the post says.

With its video editing features, Movie Gen can add, remove or replace elements as well as change the background or style depending on the post. It only targets the relevant pixels, otherwise the original content is preserved.

The model can also generate up to 45 seconds of audio based on a video and text prompts, and add ambient noise, sound effects and background music that syncs with the video content, the post says.

According to the post, possible future uses for this technology include editing videos for sharing on Reels and creating animated birthday greetings to send via WhatsApp.

“As we continue to improve our models and move toward a possible future release, we will work closely with filmmakers and creators to incorporate their feedback,” the post reads. “Through a collaborative approach, we want to ensure we create tools that help people increase their inherent creativity in ways they may never have thought possible.”