Introduction
In a groundbreaking thread on Twitter, PJ Ace (@PJaccetturo) shared the detailed process of creating a 25-minute AI-generated TV show titled "Jonah" in just three months. This blog post unrolls the thread, providing insights into the steps, tools, and challenges involved in making your own AI TV show.
The thread begins with a captivating video snippet from the show, followed by a comprehensive guide on how to replicate the process. Let's dive into the details.
We made a 25-minute FULLY AI TV show in 3 months, and I’m sharing all our secrets for free 👀
See EXACTLY how we brought Jonah’s story to life.
Here's how to create a full TV show in 15 easy steps.
Copy my PROMPTS and process below 👇🏼🧵
If you told my 13-year-old self I’d be making a biblical epic with just a computer and a few friends, I’d say that’s the coolest job ever.
The tech isn’t fully lifelike yet, but it’s wild how far we’ve come and what’s already possible.
Hope you learn from our process 🙌🏼👇🏼
Step-by-Step Process
PJ Ace outlines 15 steps to create an AI TV show, each with detailed explanations and examples. Here’s a summary of the process:
Step 1: Script.
We collaborated with Kingstone Comics to craft a 42-scene screenplay that remains faithful to the Bible while also working cinematically.
It was built as a pilot episode, structured, visually appealing, and tailored for AI production.
Step 2: Organization
We used Figma as the central hub, housing all research, style refs, and image generations.
A shared Google Sheet tracked the production stage per scene.
From script to final frame, this provided us with a bird’s-eye view, fast and clean.
The thread continues with steps 3 to 15, covering historical research, character development, visual style, shotlisting, custom GPT training, image generation, animation, consistency checks, voice performance, lip-sync, scoring, and editing. Each step is meticulously detailed, providing a blueprint for aspiring AI TV show creators.
Challenges and Lessons Learned
PJ Ace also shares the challenges faced during the production, such as consistency issues and the rapid evolution of AI tools. The thread concludes with acknowledgments to the team and tools used, and a call to action for readers to watch the full episode and join future projects.
What we learned:
– Tools evolve too fast. Midway through, Veo 3 was launched, making earlier shots feel outdated. We upgraded, but couldn’t redo everything.
– Consistency needs a dedicated role.
– We’d now use @runwayml References for character continuity, and Act II for performances
- This is why I think it's better to focus on short content right now, the longer your AI content is, the more the tools will change halfway through production.
Conclusion
This thread by PJ Ace is a treasure trove for anyone interested in the intersection of AI and television production. It demonstrates the potential of AI to revolutionize content creation while highlighting the practical steps and challenges involved.
For more details and to watch the full episode, visit the source thread on Twitter.