“We’ve seen a big explosion in image generation models, I truly believe that 2023 is going to be the year of video”, says Cristobal Valenzuela, Runway’s CEO.
Startup Runway AI Inc. is a New York-based tech company, founded in 2018, known for developing video editing software. The company, which is also a co-founder of Stable Diffusion, has released an Artificial Intelligence model that can generate videos from text descriptions in just three seconds.
The developed AI model is intelligent enough to create a short video clip from a text description such as ‘turtle flying in the sky.’
Runway’s AI text-to-video generator is taking after Meta and Google tech companies. Last year, the giant tech companies produced research papers on text-to-video AI technology.
Meta Group and Google also developed a similar AI technology that can generate short video clips from descriptive texts. Similar to Dreamix, a Google AI-powered video generative product that can redesign existing video to new graphics and style.
Runway video generator is likely to have more users than that of Meta and Google since it would be available for personal use. For now, Runway is yet to release the AI model for public use, as it considers the implications on safety and operations, but Cristobal has stated that the AI model would only be accessible for public use via Runway’s website and Discord.
The AI text-video generator uses Gen-2, which comes with video editing features, unlike the Gen-1 model earlier introduced in February, which uses text prompts to add graphics features to existing videos. Gen-2 also generates longer video clips than the Gen-1 model.
Runway believes that the AI model would aid film producers and video makers in video creation. Moreover, Runway gained a laudable level of acceptance even in its first year; in 2018, Runway AI software was used in the graphics and visual effects of movies like ‘The Late Show with Stephen’ and ‘Everything Everywhere All at Once.’
Runway AI video editing software has also gained acceptance among users of social media platforms such as TikTok and YouTube. According to Cristobal, we are getting close to the point where most online content will be auto-generated.
If most videos on the web become auto-generated without a way to differentiate real videos, what will be the possibility of getting real-life videos online? Do you think the advancement of this technology would pose a threat to online video authenticity?
Join the conversation; follow us on Facebook, Instagram, and Twitter at GoSpeedHub.
Photo by Runway
