We have not yet recovered from the advent of DALL-E2, Midjourney and companions that Meta announced Make A Video, a tool that generates short video clips from text descriptions. It's the next step in the world of AI-generated content.
This is the first time that a text to video converter tool has come this close to final launch. "Artificial intelligence research is pushing creative expression forward by providing people with tools to create new content quickly and easily," reads Press release presentation.
Make-A-Video is able to bring creativity to life with a few words or lines of text and to create distinctive films rich in colors, characters and settings. The system can also transform existing photographs or videos into similar new movies.
Great shot, David
"It is much more difficult to generate videos than photos," says Meta's CEO Mark Zuckerberg in a post on Facebook. But go, I didn't think. "In addition to correctly generating each pixel, the system also needs to predict how they will change over time. Make-A-Video solves this problem by adding an unsupervised learning layer that allows the system to understand motion in the physical world and apply it to the traditional generation from. text to image ".
Il website from Make-A-Video features some example videos made by artificial intelligence, such as "a dog wearing a superhero costume with a red cape flying in the sky" and "a teddy bear painting". It is yet another demonstration of the incredibly rapid progress of these systems. Only two? Three years ago? These things were practically science fiction.
Make-A-Video, wonder (and of course dangers)
As we increasingly rely on AI to generate art, it will be increasingly important for companies to adopt transparency policies on these algorithms. Reading the research paper behind Make-A-Video, it appears that this artificial intelligence has been "trained" using a subset of a dataset called LAOIN, which also includes unclean images. Which? ISIS executions, non-consensual nudes and so on. Meta ensures that it has sifted through this data well, automatically discarding nude and other fake images.
Sara. Meanwhile, the battle over ethics continues.
The introduction of text-to-video as a tool for artists and creators also complicates the (already very thorny) question of the legitimacy of AI-generated art. In August, you know, a guy named Jason Allen won an art competition using an image created by Midjourney, sparking a hive of controversy.
Companies that collect images for commercial use (such as Shutterstock or Getty Images) have also closed the door to this content. No ethical question, in this case. Legal only. Who is the owner of the images used by the algorithms to train? Is turning those images into new things a copyright infringement or not? The laws have not yet adapted.
Meanwhile, the tsunami continues: these technologies are literally overwhelming the public, with the same speed they show in learning to perfect. Yesterday's announcement on Make-A-Video follows by just one day on public release of DALLE-2 by OpenAI. The company that developed DALLE-2 removed the waiting list from the system, allowing anyone to generate images from lines of text.
But even as the public has access to more and more AI art-generation tools, some of the fundamental ethical questions about their use remain open: and they need to be answered.