I've been heavily into AI design now for close to two years, and so much is changing. I often ignored AI video generation, not impressed by the final results. But now, it's changing... Veo has taken the next step in quality, consistency, and control.
I started practicing using Veo via the Google Whisk generator, an easy way to generate video, especially animated character styles like Pixar. I decided to try out the 2-month pro level for $10/month (which allows 100 8-second videos per month). Control and prompt cohesion are still major issues; I'd say 60% of the videos I generated were complete failures—not following what my prompts were instructing.
Understand, we're not talking about major animations here at this point. For the most part, like AI image generation, interactions between several characters are still difficult. Just getting characters to look at each other is sometimes like pulling teeth. Asking for too much in the prompt often results in the AI ignoring everything. Again, when it works...
This second video shocked me—it added every character I requested in the background and more. Udio provided a short Halloween tune, and ElevenLabs provided vocals and sound effects. It's not long nor totally groundbreaking—other than the fact that this took me less than two hours to create a near-Pixar-level group of characters for a 15-second clip.
At this rate, I expect to generate short cartoons of my storybook transcripts by 2027. If I were to try to do this video via 3D models, recording audio, and playing my instruments... I probably wouldn't finish it due to the time it would have taken. Those who are opposed to AI... I understand.
But...
like how Photoshop killed mechanical artist jobs and darkroom photography editing,
like how computers took over how animation is done,
and when music editing went totally digital...
I am embracing this new tool A.I. -- to get my ideas down + to dream of creating things I never thought I'd have the chance to do in my lifetime.