A short YouTube segment by Matt Wolfe evaluating AI video generators for creating viral social videos. He compares tools like Invideo, Sora 2, Leonardo (using VO3), and the 'cling' model, describing mixed results—some generated footage mismatches the script and visuals are often unsatisfactory; he questions whether AI workflows are faster than manually sourcing B-roll.
These guys are kind of killing it right now. These things are super viral, like auto tuned songs with lessons that feel straight out of a high school textbook. So I want to make one of these accounts for myself and start raking it in. So jumping into Invideo, I chose TikTok stock media and then asked it to keep the total video length to 90 seconds.
Air enters through the intake valve. Fuel is sprayed into the cylinder. A piston moves downward, drawing the mixture in. The intake valve closes, the piston rises, compressing air and fuel tightly.
Some of this footage has absolutely nothing, nothing to do with what's being said in the script. Now the absolute best video generator of the moment is Sora 2. Now, I'm not totally getting what I want. This one is, you know, almost correct. Maybe I can get these images animated for me. So I'm going to use Leonardo to generate with VO3. And just for fun, let's have it generate with the cling model as well. Here's the two sets of visuals that I got. Neither of them are very good. Go ahead. I just can't get the visuals to look the way I want them. There are ways to do it with AI. Whether those ways are faster than just sourcing B roll right now, I don't think so.
AI News: Dots, GPT-6.1 Sol, Sonnet 5.5, Gemini 4, and everything you need to know
