Meta (META.O) announced on Friday the launch of a new AI model called Movie Gen, capable of generating realistic video and audio clips based on user prompts. The company claims that Movie Gen rivals tools from leading media generation startups like OpenAI and ElevenLabs. Meta showcased samples, including videos of animals swimming and surfing, as well as clips where real people's photos were used to depict actions like painting.
In addition to video creation, Movie Gen can generate synchronized background music and sound effects. The tool can also edit existing videos, such as adding pom-poms to a man running in the desert or transforming a dry parking lot into a puddle-filled scene for a skateboarding video.
According to Meta, Movie Gen can produce videos up to 16 seconds long and audio clips up to 45 seconds. The company shared data from blind tests showing that Movie Gen performs favorably against competitors like Runway, OpenAI, ElevenLabs, and Kling.
This announcement comes as Hollywood grapples with generative AI technology. Earlier this year, Microsoft-backed OpenAI demonstrated its own tool, Sora, capable of creating film-like videos from text prompts. While some technologists in the entertainment industry are eager to embrace AI to streamline filmmaking, others are concerned about the use of systems trained on copyrighted works without consent. Additionally, lawmakers have raised alarms about the use of AI-generated deepfakes in elections worldwide, including in the U.S., Pakistan, India, and Indonesia.