A former animator from the famous Pixar studios believes that, for the moment, tools capable of generating videos using AI are not yet relevant enough. The need to rework plans prevents these programs from breaking into the world of cinema.
When OpenAI introduced Sora, a new AI-powered tool that converts sentences into 60-second videos, last February, many worried about the future of animated films. Are the new programs from Silicon Valley likely to replace cinema designers? Not yet according to a former Pixar animator. In a video posted by a student at California College of the Arts, Craig Good, who worked on classics such as Toy Story and Finding Nemo, believes that generative AI tools currently have one major flaw: it is difficult to return to “the work”.
Advertisement
“ If I try to use this program in a production context, my first question will be, how can I review it? If I tell him that ultimately I hate what's happening with the background, can we do the exact same thing again, but with a fixed or different background ? », Declares Craig Good in the video, spotted by the media Gizmodo.
A long work of revising shots in animated films
Artificial intelligence tools allow users to create images and videos based on a few instructions, but none of these programs currently allows the image to be meaningfully reworked. Midjourney includes correction functions on its images, but this remains very limited and cannot lead to major changes.
“ I spent decades at Pixar making small changes to plans. The director will give fairly specific notes that the animator, the artist will have to interpret, then show that revised work the next day and then receive further notes on it. I don't know how you could use it in production if you can't constantly rework in a controlled manner » explains Craig Good.
However, the host wants to emphasize that things could change. Advances in AI remind him of the first developments in computer graphics in the 1980s. We are only at the beginning of the journey.
Advertisement