Meta Launches AI Tool To Compete With OpenAI's Video, Audio Tools

Meta has introduced Movie Gen, a new artificial intelligence (AI) model capable of generating highly realistic video and audio clips from user prompts, potentially transforming media production. 

The tech giant claims that Movie Gen outperforms tools developed by leading AI startups like OpenAI, ElevenLabs, and Runway, bringing Meta into the competitive space of generative media technology. The model allows users to create videos up to 16 seconds long and background audio for up to 45 seconds, while offering tools to edit existing video content.

Some of Movie Gen’s early demonstrations include videos of animals swimming and surfing, as well as the ability to modify existing footage. In one instance, Meta showed the tool inserting pom-poms into the hands of a runner in the desert, while in another clip, it transformed a dry skateboarding area into a water-filled parking lot. The AI can even synchronise sound effects and background music to the visual content, making it a comprehensive solution for video generation and editing.

While the tool's potential applications in entertainment are vast, Meta has taken a cautious approach regarding public release. Unlike its Llama series of language models, Movie Gen will not be made openly available to developers at this time. Instead, Meta is working closely with the entertainment industry and content creators, with plans to incorporate Movie Gen into its own products in 2025. This decision reflects ongoing concerns about the misuse of generative AI technology, especially for creating deepfakes or other misleading content in sensitive contexts such as elections.

The debut of Movie Gen comes amid broader debates about the use of AI in media, particularly in Hollywood. Microsoft's OpenAI caused a stir earlier this year with its own generative video tool, Sora, which raised ethical concerns about potential copyright infringement. Actor Scarlett Johansson publicly criticised OpenAI for allegedly imitating her voice without consent, highlighting the legal challenges surrounding AI-generated content. 

Major film studios like Lions Gate have begun experimenting with AI, giving startups access to their content libraries for training purposes in exchange for tools that can aid in film and television production.