
Midjourney has officially introduced its first AI-driven video generation model, enabling users to create short animated videos from images uploaded or created on the platform. Accessible via the web and through Midjourney's Discord server, this new feature is currently available to users with a paid subscription. The initial version allows for the generation of five-second animated clips. After an image is generated, users will notice an 'animate' button that guides them through a prompt-based animation process. By default, the system applies a generic motion prompt, but users have the option to provide custom descriptions for the movement. Additionally, users can specify a starting image to influence the animation. Midjourney's animation capabilities can be extended in increments of four seconds, up to four times, allowing for videos as long as 21 seconds. The platform offers both high and low motion modes, giving users the flexibility to choose whether the subject, camera, or both elements are animated. The pricing model is based on GPU usage, with subscriptions starting at $10 per month, which includes 3.3 hours of fast GPU time—equivalent to generating around 200 images. For video creation, it is estimated that costs will be approximately eight times higher than that of a single image, translating to the expense of one image per second of video. In a post announcing this innovative feature, Midjourney founder David Holz described it as a 'stepping stone' towards more sophisticated models that could facilitate real-time open-world simulations in the future. However, this launch comes amid ongoing legal challenges, as Midjourney faces a lawsuit from Disney and Universal. The studios have raised concerns regarding the company's video ambitions, labeling it a 'virtual vending machine' for unauthorized reproductions of copyrighted material, particularly focusing on the potential intellectual property infringements associated with its video generator. Midjourney's entry into the AI video generation arena aligns it with other tech giants like OpenAI, Google, and Meta, all of whom are racing to develop advanced content creation tools that transform text prompts into engaging videos.
During his keynote at the annual GTC event, Nvidia CEO Jensen Huang captivated audiences for over two hours, showcasing ...
TechCrunch | Mar 21, 2026, 16:45
At this year's Oscar celebration with friends, I decided to take a unique approach: I let Anthropic's Claude AI predict ...
Business Insider | Mar 21, 2026, 23:50Mark Zuckerberg's vision of a metaverse, which he championed nearly five years ago, has faced significant scrutiny as re...
Business Insider | Mar 21, 2026, 09:25Hachette Book Group has decided to withdraw the upcoming release of the novel "Shy Girl" due to concerns that artificial...
TechCrunch | Mar 21, 2026, 19:40
In 2021, Facebook rebranded itself as 'Meta' to align with its vision for the metaverse. However, this move has become e...
Business Insider | Mar 20, 2026, 23:10