Meta (the parent company of Facebook, Instagram, WhatsApp, and Threads) has announced the development of a new artificial intelligence model called Movie Gen. This model can generate realistic video and audio clips based on simple user commands, with Meta claiming it can compete with AI tools from top companies like OpenAI and ElevenLabs.

Sample creations provided by Meta included videos of animals swimming and surfing, as well as clips using real photos of people, depicting them performing actions such as painting on a canvas.

According to Meta, Movie Gen can also generate background music and sound effects synchronized with video content, as well as edit existing videos using the tool.

The videos created by Movie Gen can last up to 16 seconds, while the audio clips can be as long as 45 seconds. Meta also presented results from blind tests showing that its model outperforms those from other companies.

The announcement comes as Hollywood grapples with how to incorporate generative AI into visual effects production. In February, OpenAI demonstrated how its Sora tool could generate video resembling feature-length films in response to text prompts.

Technologists in the entertainment industry are eager to use such tools to enhance and accelerate film creation, while others worry about adopting systems that may have been trained on copyrighted works without permission.

Source