This AI creates a video from your words

It should be obvious to anyone following the generative AI news that the new technology is rapidly becoming ubiquitous. Last year, AI image generators took the world by storm, and towards the end of 2022, ChatGPT captured the public imagination. Now, a few weeks into 2023, text-to-video AI is almost here.

This week, AI startup Runway announced Gen-2, a multimodal AI system that can generate videos with clips of other videos, images, or even just text. That's right, before very long users will be able to type anything they want into a prompt and get a fully produced video of what they imagined in their imagination.

Generate videos with nothing but words. If you can tell it, now you can see it.

Introducing Text to Video. With Gen-2.

Learn more at https://t.co/PsJh664G0Q pic.twitter.com/6qEgcZ9QV4

— Runway (@runwayml) March 20, 2023

However, as with all new technology, it's not quite ready for prime time. According Gizmodo's Kyle Bar, while Runway's new video AI is not yet available to the public, there is already another video synthesis service: ModelScope, which was released just a few days ago. Its website is mainly in Chinese, with some sections in English. But the sample AI-generated videos on the site are pretty impressive, if crude.

Some of the examples on the ModelScope site include "a giraffe under a microwave", "a goldendoodle playing in a lakeside park", "a panda driving a car", "a teddy bear running in New York", etc. . Each video clip is only a few seconds long but clearly demonstrates the power of the new technology. It's worth noting that each sample video contains a Shutterstock watermark, likely because the company used stock footage to train its AI.

However, ModelScope is not exactly user-friendly. In addition to the site being mostly in Chinese, it seems users need to do a bit of research (or know the ins and outs of generative AI) to make it work. It's not like ChatGPT or the New Bing yet. But the very existence of this technology on the internet means that text to video is coming much sooner than many of us thought.

Source: Gizmodo