Photo by Wallace Silva on Pexels

Midjourney, renowned for its AI image generation, has officially entered the video arena with the release of its first AI video generation model. The initial iteration focuses on creating short, looping videos from user-provided or AI-generated images, paving the way for more advanced features in the future.

After generating an image within Midjourney, users will now find an “animate” button, which allows them to produce a roughly 5-second video clip. This can be based on either a text prompt describing the desired motion or an uploaded image. While the tool defaults to a basic “make things move” animation, users have the flexibility to customize the motion and direction. Furthermore, the duration of these animations can be extended up to 21 seconds, and users can adjust the intensity of the movement.

Currently, the AI video tool is accessible via Midjourney’s website and Discord server. Access requires a Midjourney subscription, which starts at $10 per month, and video generation jobs are priced higher than image generation. This launch comes amidst an ongoing copyright lawsuit filed by Disney and Universal, who have expressed concerns regarding the potential for copyright infringement with AI-generated video content.

Midjourney founder David Holz has characterized this release as a “stepping stone” towards the development of real-time, interactive open-world simulations, hinting at the company’s long-term ambitions in the AI video space. The launch places Midjourney alongside tech giants such as Google, OpenAI, and Meta, all of whom have also recently debuted their own AI video generation platforms.