It’s probably not the same level as watching the Space Race…
… But the pursuit of AI video generation sure is interesting to watch.
What’s going on: Days after Meta announced Make-a-Video—its artificial intelligence (AI) video generator—Google and Phenaki announced video generators of their own.
Looks good: Google’s Imagen generator can produce high definition videos several seconds long.
Yes, initial images still appear distorted, and objects and movement can look unnatural and blend in impossible ways… but the results are still remarkable.
Story time: Also, the founders of Phenaki launched their own text-to-video generator yesterday.
1/ Today we are excited to introduce Phenaki: https://t.co/7xkcoeuXwB, short-link-to-paper, a model for generating videos from text, with prompts that can change over time, and that is able to generate videos that can be as long as multiple minutes! pic.twitter.com/icO8M0frKk
— Dumitru Erhan 🇺🇦 (@doomie) October 5, 2022
Phenaki prompts allow room for narratives and stories, and can generate videos lasting several minutes. Wild.
Why we care: It seemed impossible a few years ago, but AI-produced video is now becoming a viable industry with multiple competitors.
These tools could make video production much easier—and cheaper— for marketers without the skills or budget.