Runway Launches ‘Expand Video’, Turning Clips into Cinematic Masterpieces

Runway has unveiled an innovative new feature called ‘Expand Video’ which allows users to extend video frames while maintaining visual consistency, enabling new creative and cinematic possibilities. The Expand Video feature is being gradually rolled out and will soon be available to all users as part of the Gen-3 Alpha Turbo suite.

“Fridays are for releasing,” said Runway CEO Cristóbal Valenzuela on X while announcing this new feature. Runway completed six years last week. 

This tool supports text prompts and guiding images, enabling creative expansions into new directions. Users can chain multiple expansions to produce cinematic movements such as crash zooms and pull-out reveals from static footage. Comprehensive tutorials are available at Runway Academy to help users leverage these features effectively.

On the recent viral conversation about AI’s impact on Hollywood involving academy winner Ben Affleck, Valenzuela said: “The history of cinema is the history of technology. The history of art is the history of technology.” 

Runway Making Big Moves

In October, Runway introduced Act-One, a tool within Gen-3 Alpha for generating expressive character performances.  Before that, Runway partnered with Lionsgate, the studio behind John Wick, granting the studio exclusive access to Runway’s AI tools for content creation. This collaboration set a new standard in the entertainment industry and positioned Runway ahead of competitors like OpenAI’s Sora. 

There are rumours about a potential AI collaboration between Disney and OpenAI. Similarly, Promise, a studio backed by Andreessen Horowitz, will use generative AI to produce shows and movies. 

Runway’s tools also played a role in the Oscar-winning film Everything Everywhere All at Once, streamlining special effects production, reducing costs, and minimising manual effort.

“Apparently rumor is that OpenAI is planning to finally release Sora in ~2 weeks. What I dread the most is having to listen to agi people go on podcasts to talk about the future of art and creativity,” said Valenzuela referencing Sam Altman’s recent interview predicting AGI could arrive by 2025.

Interestingly, while Runway has been shipping features, there is no update on the official launch of Sora since its announcement in earlier this year. There were rumours of its launch after the elections. 

In terms of competition, China has become a major competitor in AI video tools, with platforms like Kuaishou’s Kling surpassing others like Sora. Kling offers realistic motion simulations, 3D face and body reconstructions, and user-trained AI characters, gaining praise for its advanced features. MiniMax, another Chinese text-to-video generator, has also been recognised for top-tier AI videos. While tools like Kling, Sora, and Luma’s Dream Machine remain limited in access, Singapore-based Pollo AI by HIX.AI aims to democratise AI video generation.

(Total 15 views)