Midjourney Debuts AI Video Model Amid Disney Lawsuit
▼ Summary
– Midjourney launched its first AI video generation model, V1, allowing users to animate images into 5-20 second clips with text guidance.
– The new video tool is integrated into Midjourney’s existing subscription plans, starting at $10/month, with costs comparable to image generation.
– Midjourney’s video model lacks sound and advanced editing features compared to competitors like Runway and Luma AI.
– The company faces a major lawsuit from Disney and Universal over alleged copyright infringement related to AI-generated content.
– Midjourney aims to develop a unified “world model” for real-time interactive environments, positioning itself in a competitive AI video market.
Midjourney has taken a bold step into AI video generation with its new V1 model, allowing users to transform static images into short animated clips. This move positions the company alongside major players in the rapidly evolving AI video space while simultaneously facing legal scrutiny from entertainment giants Disney and Universal.
The platform enables users to animate images, either generated within Midjourney or uploaded externally, into 5-second clips, with options to extend sequences up to 20 seconds. Two motion modes are available: low motion for subtle animations like blinking or gentle environmental shifts, and high motion for more dynamic movement, though the latter may introduce visual artifacts. Each animation job produces four variations, giving creators multiple options to choose from.
Pricing remains competitive, with video generation included in existing subscription plans starting at $10 per month. While each video job costs roughly eight times more than an image generation, the per-second cost aligns with Midjourney’s still-image pricing, making it an affordable option for short-form experimentation. A “video relax mode” is also in testing for Pro-tier subscribers, offering slower processing in exchange for reduced costs.
However, Midjourney’s video model has notable limitations. Unlike competitors such as Google’s Veo 3 or Luma Labs’ Dream Machine, it doesn’t generate sound, requiring users to add audio separately. The platform also lacks advanced editing features like timeline adjustments or scene transitions, focusing instead on simplicity and accessibility.
The launch comes amid intensifying competition in AI video generation. Chinese startup MiniMax recently unveiled Hailuo 02, praised for its realism and cinematic motion, while Luma Labs and Runway offer features like audio co-generation and video restyling. Meanwhile, OpenAI’s Sora and Google’s Veo are pushing toward multimodal synthesis, integrating text, visuals, and sound into cohesive scenes.
Legal challenges loom large for Midjourney, as Disney and Universal have filed a copyright infringement lawsuit accusing the company of training its models on protected characters like Darth Vader, Elsa, and Shrek. The lawsuit also preemptively targets Midjourney’s video service, alleging it enables derivative content without proper licensing. This case could set important precedents for AI-generated media and copyright law.
Looking ahead, Midjourney aims to develop a “world model”, a unified system combining static images, motion, 3D navigation, and real-time rendering. This vision aligns with broader industry trends, as companies like Runway, DeepMind, and Microsoft explore AI-driven simulations for gaming, virtual environments, and enterprise applications.
For now, Midjourney’s video tool offers an accessible entry point for creators, but its long-term success hinges on navigating legal hurdles and keeping pace with rivals in an increasingly crowded market. Enterprises weighing adoption must consider both its affordability and potential legal risks before integrating it into workflows.
(Source: VentureBeat)
