Have you seen all those crazy AI videos taking over social media, like Bigfoot vlogging or Stormtroopers in disguise, and wondered how they’re made? I know I have! Well, I just stumbled upon an incredible masterclass from an AI professional who breaks down exactly how to use Google’s Veo 3, and it’s a total game-changer.
This expert demystifies the whole process, from getting access to pushing the tool to its absolute limits.
⚙️ How to Get Your Hands on Veo 3
First off, how do you even use this thing? The creator laid out a few solid options:
- Google’s Free Trial: You can get a free one-month trial of the Google AI Pro plan, which includes Veo 3.
- Perplexity on X: The YouTuber mentioned you can try using the
@AskPerplexitybot on X (formerly Twitter) for free generations, but warned it can be a bit hit-or-miss. - Third-Party Tools: Platforms like Krea AI or Leonardo AI have Veo 3 built-in. This industry pro found that Leonardo AI was one of the cheapest paid options for getting consistent access.
✍️ Recreating Viral Memes
This is where it gets fun. The creator showed how to recreate some of the most popular AI memes. The secret? Detailed prompts!
For the Stormtrooper and Bigfoot vlogs, this innovator actually used ChatGPT to improve his initial ideas. He’d give ChatGPT a simple prompt and ask it to “add more detail including both visuals and audio.” The results were awesome and looked just like the viral clips we’ve all seen.
✨ The Secret to Consistent Characters
Getting the same character to appear across multiple video clips is the holy grail of AI video, and this channel shared a few brilliant methods to pull it off:
📌 Image-to-Video: Start with a generated image of your character. The YouTuber created a fictional punk band called “Norc” and used that one image to generate multiple performance clips. Since the AI starts with the same source, the band members stayed consistent.
📌 Google Flow: This tool, which comes with the Google AI plan, is built for storytelling. The creator showed how you can take an initial video and use a feature called “jump to” to create a new scene with the same characters in a totally different environment, like playing on an active volcano!
📌 The ChatGPT Combo: He also demonstrated a clever workflow where he’d give ChatGPT an image of his band and ask it to place them in a new setting, like “on the Titanic as it’s sinking.” He then took that new image generated by ChatGPT and used it as the starting frame in Veo 3 to animate the scene.
🚀 Pushing Veo 3 to its Limits
To see what Veo 3 would and wouldn’t do, the person who shared it tested some truly wild prompts. While it successfully created talking hamburgers on a grill and a Space Marine cuddling a bunny, it failed when asked to generate a:
“human-sized hot dog cooking human-sized humans.”
So, yes, there are limits!
Interestingly, the YouTuber found he could get away with more edgy or bizarre prompts using third-party APIs like Leonardo’s compared to Google’s native interface.
This was an amazing deep dive into what’s possible right now with AI video. The mind behind it even shared an alternative tool for when Veo 3 falls short.
For the full deep-dive and to see all the hilarious video results for yourself, make sure to watch the original video from the creator!