AI company Midjourney has its first video model. This first take of AI-generated video allows users to animate images. This is either made with Midjourney or uploaded from another source. The initial result is a 5-second clip that allows the user to choose to extend it up to 4 seconds. For now, the video can only be generated on the web, and requires a subscription of at least $10 a month to access it.
Midjourney was one of the early names in the space of AI-generated still images, despite other platforms pushing the forefront of debate over artificial intelligence development. Google's latest I/O conference includes several new tools for AI-generated videos. This includes text-to-video and tools for filmmakers. Openai's SORA was also a text-to-video option last year, but Adobe's recent ability to create videos from text or image prompts.
But being a little behind video games doesn't stop them from portraying the rage of creatives who claim that the model is illegally trained. In fact, the announcement of this video follows Hot shortly after the lawsuit against the company. Last week, Disney and NBCuniversal Midjourney said they had allegedly infringed copyright. And, like AI tools, it can always be misused. But Midjourney asked people “use these technologies responsibly” and that nothing is going to work.
