The Newest Text to Video AI Is Here, I've Got the Reviews—and the Breakdown On Their All-Consuming TOS

Within the span of 48 hours two new text-to-video models just dropped: one paid, one unpaid. The paid one is the only one (sort of?) worth a damn—so here's how to get the most out of the free trial.

Mind-blowing that this is an AI image. The prompt is after the jump, and I’ve used it as the training base to make a 4-second “explosion” video.

Text-to-video AI generation still has a long, long way to go but it’s remarkable what it can do so far.

If it’s anything like the Midjourney trajectory of improvement then in a span of months it’ll be stunning.

I’ll also give you the Beijing-based Hugging Face model that’s responsible for all those viral Conan O’Brien eats fried chicken then crashes a car/ Will Smith eats spaghetti AI videos that being going around for months (and even made the late-night talk shows).

Plus the most no-fluff, no-hype, only-value AI YouTuber I’ve found so far.

Want to see cats playing a symphony, teddy bears disco dancing, a puppy playing guitar and Darth Vader doing water ballet?

I got you.