Hello from TIMEWELL
I'm Hamamoto from TIMEWELL Inc.
Looking for AI training and consulting?
Learn about WARP training programs and consulting services in our materials.
AI Tools Are Revolutionizing Our Creative Processes
In the digital age, AI tools are revolutionizing our creative processes. This article introduces a workflow for creating four-panel manga by combining two cutting-edge tools — ChatGPT and NanoBanana — together with Canva.
In the past, bringing a detailed story to life through an illustrator required significant time and cost. The latest AI technologies have dramatically lowered that barrier. This article draws on the specific workflow demonstrated in a video to explain in detail how AI tools can supercharge creative work. The possibilities extend to corporate training materials, recruiting websites, and internal communication tools.
We walk through every step in detail — from story generation to image creation to final editing in Canva. By the end of this article, your creative work will be dramatically more efficient, and you will be able to achieve more compelling visual communication than ever before.
[How ChatGPT Generates the Four-Panel Manga Script] [High-Quality Image Generation with NanoBanana and Its Practical Applications] [Perfecting the Final Product: Practical Canva Editing Techniques] [Summary]
How ChatGPT Generates the Four-Panel Manga Script
The workflow introduced here uses AI tools to supercharge every stage of four-panel manga creation — a process that previously required outsourcing to skilled illustrators. Traditionally, the hardest step in manga production was deciding on an overall story structure and figuring out how to arrange scenes within each panel. For four-panel manga in particular, the challenge was fitting a complete narrative arc — setup, development, twist, and resolution — into just four frames.
In this workflow, a purpose-built custom ChatGPT (GPTs) is used so that users simply input a rough theme and direction, and the system automatically generates a detailed storyline. In the demonstration, for example, entering the theme of new graduate recruiting along with direction and target audience settings produced a complete output that included the title, panel-by-panel content (scene setup, dialogue, narration, character actions and expressions), character names, and even specific prop ideas.
The most impressive aspect of this process is how GPTs understands the user's input keywords, general direction, and target readership — and uses that information to generate a logical, original story. By simply entering the theme "new graduate recruiting," users receive a fully developed script that includes character relationships (a department manager, a nervous new graduate, and a colleague named Sato) and the complete story progression. Because this prompt input follows instructions built into the specialized GPTs, it delivers higher-quality output than would typically be possible with standard ChatGPT.
The Script Goes Far Beyond Just Text
The script goes far beyond just text — it covers panel composition, production notes, and psychological characterization for each scene. For example, an interview scene might include detailed instructions such as "Tanaka is visibly nervous" and "the manager smiles, relaxed, and serves tea," along with the movement of characters, their expressions, and even a description of cherry blossoms falling in the background. This level of detailed description is essential for enhancing the visual impact of the manga and leaving a strong impression on readers. The prompt also includes mechanisms to maintain character consistency and story continuity, resulting in thoughtful and polished four-panel manga.
The key points at this stage can be summed up as follows:
By inputting a rough theme and direction, ChatGPT (GPTs) automatically generates the storyline
The output includes character names, scene details, dialogue, and production notes
After Prompt Input, the Script Is Ready Almost Immediately
After prompt input, the script is ready almost immediately, eliminating the need to outsource
Through this prompt input process, users can complete manga script creation in-house — quickly and cost-effectively — without the outsourcing that used to be required. Scene instructions even include animation-style guidance on movement and atmosphere, enabling a more vivid manga world to come to life.
This GPTs is also designed to output in a format optimized for image generation in NanoBanana, which further contributes to workflow efficiency. Moreover, this story-generation GPTs works effectively even in the free version of ChatGPT, making it possible to create high-quality content while keeping initial costs low. This gives both companies and individual creators the environment they need to quickly bring their ideas to life.
Instructions for Users Are Refreshingly Simple
Instructions for users are refreshingly simple — you only need to describe the theme, character setup, and a rough scene progression — making it accessible even for beginners. As a result, it is possible to arrive at a finished four-panel manga much faster than with traditional storyboarding or panel layout work. The ability to create stories suited to a wide range of industries and needs — including entertaining narratives for business scenarios like new graduate recruiting and employee training — is not to be overlooked.
This process also makes a significant contribution to reducing outsourcing costs, delivering tangible benefits in both effort and expense. In today's fast-paced business environment, this underscores the growing importance of AI-powered creative production.
[High-Quality Image Generation with NanoBanana and Its Practical Applications]
Once the story framework is in place, the next step is generating images to bring it to life visually. This is where NanoBanana — an advanced image generation tool — comes in. NanoBanana uses the image generation capabilities embedded in Gemini to understand each scene of the script created by ChatGPT and quickly produce high-quality images featuring anime-style characters and backgrounds. NanoBanana's greatest strength lies in its contextual understanding of prompts and the quality of its animation-style imagery.
Image Generation in NanoBanana
Image generation in NanoBanana begins with the user copying the entire story script output by ChatGPT (GPTs) and pasting it into NanoBanana as a prompt. Care is taken to include each panel's dialogue, production notes, and background details in the prompt, and the tool generates images that match it as accurately as possible. The output may sometimes be a six-panel manga rather than four panels, but further prompts can be added to make adjustments. During image generation, fine settings are applied to maintain consistency in each character's face, resulting in a finished product that feels cohesive throughout.
In the actual demonstration, the tool automatically generated: panel 1 — the tense atmosphere of the interview venue; panel 2 — a miscommunication about a name badge; panel 3 — the manager's unexpectedly positive reaction; panel 4 — the final hiring decision. Each panel maintained largely consistent character expressions and clothing design thanks to NanoBanana's high-precision image generation, resulting in a visually coherent overall presentation.
However, NanoBanana does have one weakness: it struggles with generating Japanese text within images. Text portions in generated images may require manual correction or editing in Canva. That said, NanoBanana is not just about generating images — it is the key to maintaining strong continuity between story and visuals. Once images have been generated, any fine adjustments can easily be made in Canva afterward — a significant advantage. Through this process, even if the initial output is not perfect, quality can reliably be improved in subsequent steps, ultimately producing a polished four-panel manga.
This Generation Process Has Been Verified in Practice
This generation process has been verified through hands-on demonstrations, which showed that while initial image generation may have minor imperfections, subsequent corrections produced an impressive finished manga. These fine-tuned adjustments are proof of the high-quality visual content that the NanoBanana and Canva combination delivers.
[Perfecting the Final Product: Practical Canva Editing Techniques]
With the four-panel manga nearly complete after NanoBanana's image generation, the final stage involves editing in Canva, which plays a critical role. Canva is widely used as a simple yet powerful design tool, and its ease of use is on full display in this workflow. Editing in Canva covers a wide range of steps — cropping images, erasing unwanted text or details, and adding speech bubbles and text — and constitutes the decisive step for ensuring the consistency and visual appeal of the final four-panel manga.
First, users upload the images downloaded from NanoBanana to Canva. The cropping tool in Canva is used to adjust image sizes and layouts, precisely cutting out the necessary portions of each panel. For unwanted text or imperfections that arose during NanoBanana's generation process, Canva's shape tools are used to cover them up discreetly.
In the Actual Demonstration
In the actual demonstration, the first panel was edited in Canva to add a speech bubble. Through careful editing of text placement and sizing, and even converting text to vertical orientation where needed, a more readable, story-driven layout was achieved. The demonstration also showed attention to detail, such as flipping speech bubble assets to the correct orientation.
Another major aspect of Canva editing is layer management — the order in which images are stacked. By adjusting the z-order of shapes, text, and panel image layers, users can achieve a unified final layout. In practice, corrections are made by placing background shapes at the very back and layering text and speech bubbles on top as needed. Once you get used to it, this work in Canva flows very smoothly and even becomes an enjoyable creative process in itself.
Canva also offers a technique for erasing unwanted text using shape tools such as rectangles and polygons. When generated images contain unwanted text, simply overlaying an appropriately colored shape produces a natural-looking result — ultimately achieving a professional finish. When adjusting the size and position of each image, using the pink guide lines and grid features that Canva displays makes it easy to balance each panel's placement, resulting in a well-proportioned four-panel layout.
The Moment Everything Comes Together
During the Canva editing process, there is a moment when the story created in ChatGPT (GPTs) and the images from NanoBanana unite as a single visual story. Because each step links with the others and each panel's content and character settings directly impact the final quality, users must be willing to attend to every detail. The four-panel manga that emerges from this process becomes a unique piece of content that can be used for everything from new graduate recruiting to internal training. This is truly the new shape of future creative work — brought to life by the unified system of ChatGPT, NanoBanana, and Canva.
We have now walked through the complete four-panel manga creation workflow using ChatGPT (GPTs), NanoBanana, and Canva. The key point is that using a dedicated GPTs to automatically generate a story script based on a theme such as new graduate recruiting dramatically reduces the time and cost of outsourcing. NanoBanana is then used to generate high-quality images for each scene, with character depictions and scene settings carefully reproduced to maintain strong continuity with the story. In the final stage, Canva is used for trimming, text insertion, and background corrections to produce a consistent, polished four-panel manga.
This workflow eliminates much of the labor that traditional creative production demanded, providing an environment in which anyone can easily create high-quality manga content. Every step — from script to image to final visual editing — is seamlessly connected through digital tools, contributing to internal corporate communication, marketing materials, and the enhancement of individual creative expression.
The Future of Creative Work Is in Your Hands
In the years ahead, using AI tools will enable creative work to become dramatically more efficient, bringing far more ideas to life far more quickly. Please give the methods introduced in this article a try and incorporate them into your business or personal projects. The creative future is already unfolding in your hands, starting right now.
Reference: https://www.youtube.com/watch?v=82W4QFoKClg
Related Articles
- The Reality of Working Reduced Hours After Two Maternity Leaves — and How Work Perspectives Shift | TIMEWELL
- Before Paternity Leave, Part 2: Three Things You Must Do to Take Leave During a Busy Season
- A Fifth-Generation Construction Company Owner Finds His Own Path at a Hands-On Architecture Firm — Fujita Construction