AI boot corridor on Monday liberated What claims to be one of the high -faithful AI video generators.
It is called Gen-4, the model is circulating to the customers of the company and business. The corridor claims that it can create consistent characters, locations and objects in tents, maintain “coherent global environments” and regenerate elements from different angles and seats in scenes.
“Gen-4 can use visual references, in combination with instructions, to create new images and videos that use consistent style, themes, locations and many more”, Runway wrote in a blog post”[a]without the need to refine or additional training. ”
Gen-4 sets a new standard for video production and is a remarkable improvement over Gen-3 Alpha. It depends on his ability to create extremely dynamic videos with realistic motion as well as on the subject, object and style with superior attachment and the best world … pic.twitter.com/w9aco5boj7
– corridor (@Runwayml) 31 March 2025
The corridor, supported by investors such as salesforce, Google and Nvidia, offers a series of AI video tools, including video creation models such as Gen-4. It faces fierce competition in the video production industry, including Openai and Google. However, the company has struggled to diversify, by reducing the deal with an important Hollywood studio and the succession of millions of dollars to fund films using AI videos.
Runway says that Gen-4 allows users to produce consistent characters in all lighting conditions using a reference image of those characters. To create a scene, users can provide images of themes and describe the composition of the shot they want to create.
Using visual references, in combination with instructions, the Gen-4 allows you to create new images and videos in consistent style, themes, locations and more. Allowing continuity and control to your stories.
To try the model’s narrative abilities, we have gathered… pic.twitter.com/iyz2baew2u
– corridor (@Runwayml) 31 March 2025
“Gen-4 is superior to its ability to create extremely dynamic videos with realistic motion, as well as in relation to the subject, object and style with higher compliance and better understanding of the world class,” the company claims in its blog post. “Gen-4 corridor [also] It represents an important milestone in the ability of visual genetic models to simulate the real world’s physics. ”
Gen-4, like all models that create videos, has been trained in a huge number of video examples to “learn” the patterns in these videos to create synthetic materials. The corridor refuses to say where the training data came from, partly out of fear of sacrificing the competitive advantage. However, training details are also a possible source of IP pipelines.
In this case, the corridor is Facing a suit They brought from artists against her other genetic companies who accuses the accused of drawing up their models in copyright -protected artworks. The corridor argues that the Doctrine is known as fair use Pull it out of legal effects. It is not yet clear whether the company will prevail.
Bettings are somewhat high for the hallway, which is said to grow A new round of funding This would estimate the company at $ 4 billion. According to information, the corridor hopes to hit $ 300 million in annual revenue This year after starting products such as an API for videos that create videos.
However, the lawsuit against the corridor is shaking, AI video genetic tools threaten to upgrade the film and television industry as we know it. 2024 study Assigned by The Animation Guild, a union representing Hollywood animators and cartoonists, found that 75% of AI film production companies have reduced, unified or eliminated jobs after integrating technology. The study also estimates that by 2026, more than 100,000 US entertainment work will be disturbed by genetic AI.
