I Asked ChatGPT-5 to Create Comics, Then I Drew Them: An AI Prompt Engineer’s Experiment in 2025

  • by
  • 14 min read

As an AI prompt engineer with over a decade of experience working with language models, I've had a front-row seat to the rapid evolution of artificial intelligence. In 2025, we find ourselves at a fascinating juncture where AI has become an integral part of creative processes across industries. But just how far has AI come in its ability to generate creative content? To explore this question, I decided to put the latest iteration of OpenAI's language model, ChatGPT-5, to the test in a unique way – by having it generate comic ideas for me to draw.

The results were both surprising and illuminating, offering valuable insights into AI's current capabilities and limitations in creative tasks. Join me as I share my experiment and what it reveals about the state of AI-assisted content creation in 2025.

Setting Up the Experiment: Crafting the Perfect Prompt

As any experienced prompt engineer knows, the quality of AI-generated content is heavily dependent on the quality of the input. With ChatGPT-5's enhanced understanding of context and nuance, I knew I could be more sophisticated in my prompting compared to earlier models. Here's the template I developed:

Generate a 4-panel comic strip concept with the following elements:
1. A concise description of the visual content for each panel
2. Any dialogue or text that should appear in the panels
3. The overall theme or message of the comic
4. Suggested art style or visual tone (e.g. realistic, cartoony, minimalist)
5. Any specific cultural or technological references relevant to 2025

Genre: [specified genre]
Target Audience: [specified demographic]

I then varied this prompt, specifying different genres, target audiences, and sometimes adding constraints or specific elements to include. This approach allowed me to test ChatGPT-5's versatility and its ability to tailor content to specific requirements.

Comic #1: The Unexpected Time Traveler

For my first attempt, I asked ChatGPT-5 for a science fiction-themed comic targeting young adults. Here's what it produced:

Panel 1: A young scientist stands proudly next to a sleek, holographic time machine in a high-tech lab.
Scientist: "Behold! My quantum entanglement time displacer is complete!"

Panel 2: The scientist steps into the machine, which emits a soft blue glow and shimmering particles.
SFX: *VWORP* *VWORP*

Panel 3: The time machine materializes in what appears to be a lush, prehistoric jungle.
Scientist (peeking out): "Yes! I've reached the Cretaceous period!"

Panel 4: The scientist emerges fully, looking bewildered. Instead of dinosaurs, the jungle is filled with bioluminescent plants and floating, tentacled creatures. In the sky, multiple moons are visible.
Scientist: "Wait... this isn't Earth's past. It's someone else's future!"

Theme: The unpredictability of scientific discovery and the vastness of cosmic possibilities.

Art style: A blend of realistic backgrounds with slightly exaggerated, expressive character designs. Use vibrant colors for the alien world to contrast with the sterile lab environment.

2025 references: Incorporate holographic displays and quantum computing elements in the first panel's lab scene.

The Drawing Process

Translating ChatGPT-5's descriptions into visual form presented an exciting challenge. The AI had provided a solid framework, but many details were left to my interpretation. This ambiguity actually allowed for more creativity on my part.

I decided to make the time machine a sleek, translucent cylinder with intricate quantum circuitry visible inside. For the alien world, I let my imagination run wild with bioluminescent vegetation and floating, jellyfish-like creatures with fractal tentacles. I used digital painting techniques to create a vibrant, otherworldly atmosphere that contrasted sharply with the sterile lab environment of the first panel.

AI Prompt Engineer's Perspective

From a prompt engineering standpoint, this output demonstrates several key advancements in ChatGPT-5 compared to its predecessors:

  1. Coherent Narrative Arc: The AI successfully generated a story with a clear setup, progression, and twist ending within the constraints of a 4-panel format.

  2. Genre Adherence: The comic effectively incorporates science fiction elements while avoiding clichés.

  3. Visual Guidance: While not exhaustive, the AI provided more specific visual cues (e.g., "holographic time machine", "bioluminescent plants") that aided in the illustration process.

  4. Thematic Depth: The suggested theme shows a level of abstraction and philosophical consideration that was often lacking in earlier AI models.

  5. Contemporary References: The inclusion of quantum entanglement and holographic technology demonstrates ChatGPT-5's updated knowledge base, reflecting current scientific trends.

However, the description's lack of minute visual details still highlights a limitation of text-based AI models. They can generate plot and dialogue but struggle with the kind of specific visual descriptions that a professional comic artist might provide to an illustrator.

Comic #2: The Social Media Dilemma

For my next comic, I wanted to explore a more contemporary theme. I asked ChatGPT-5 for a comic about social media addiction, targeting millennials and Gen Z. Here's what it generated:

Panel 1: A person is lounging on a couch, their face illuminated by their sleek, flexible smartphone. Various app notifications pop up as holograms around them.
Caption: "Just five more minutes of scrolling..."

Panel 2: The scene zooms out to show the person still in the same position, but the room around them has changed. Plants have wilted, dishes have piled up, and the clock shows several hours have passed.
Caption: "Okay, maybe ten more minutes..."

Panel 3: The view shifts to show the person's neglected virtual reality workspace flickering with missed calls and deadlines, while their physical space is in disarray.
VR Assistant (hologram): "Reminder: You have 37 unread messages and a project due in 1 hour."
Person: "Shh, I'm building my personal brand!"

Panel 4: The person finally looks up from their phone with a shocked expression. Their augmented reality contact lenses display a summary of their screen time and neglected responsibilities.
Person: "12 hours?! But I just sat down! Is this what touched grass means?"

Theme: The insidious nature of social media addiction and its impact on real-world responsibilities and connections.

Art style: Hyper-realistic backgrounds contrasted with a slightly cartoonish character style to emphasize the disconnect between digital and physical reality.

2025 references: Include flexible/foldable smartphones, holographic notifications, augmented reality contact lenses, and advanced virtual reality workspaces.

The Drawing Process

This comic presented a unique challenge – depicting the gradual descent into digital absorption while maintaining visual interest across the panels. I used a combination of traditional and digital art techniques to create a stark contrast between the vibrant, holographic digital elements and the increasingly neglected physical environment.

The character's expression and posture subtly change across the panels, showing the physical toll of prolonged screen time. I paid special attention to the background details, using them to tell a story of increasing neglect as the character becomes more engrossed in their digital world.

AI Prompt Engineer's Perspective

ChatGPT-5's output here demonstrates several impressive capabilities:

  1. Topical Relevance: The AI has successfully created a scenario that humorously yet poignantly illustrates the issue of social media addiction, showing an understanding of current societal concerns.

  2. Technological Extrapolation: The inclusion of advanced AR/VR elements and flexible devices shows ChatGPT-5's ability to extrapolate current tech trends into plausible near-future scenarios.

  3. Visual Storytelling: The AI's suggestions for visual changes across panels (wilting plants, piling dishes) demonstrate an understanding of how to convey the passage of time and changing circumstances visually.

  4. Contextual Humor: The use of the "touched grass" meme in a futuristic setting shows ChatGPT-5's ability to blend contemporary internet culture with speculative fiction.

  5. Thematic Depth: The suggested theme captures the complexity of the issue, addressing both personal and professional impacts of social media addiction.

From a prompting perspective, this output underscores the importance of providing clear thematic guidance and target audience information. By specifying "social media addiction" and the millennial/Gen Z target, I guided ChatGPT-5 towards creating a narrative that effectively communicated this concept in a relatable way for that demographic.

Comic #3: The Empathetic AI

For my third comic, I decided to go meta and ask ChatGPT-5 to create a comic about AI itself, targeting a general adult audience. Here's what it generated:

Panel 1: A humanoid robot with soft, rounded features is standing in a living room, facing an elderly human who looks skeptical.
Robot: "Hello! I'm Nova, your new AI companion."
Human (thinking): "Great, another soulless machine."

Panel 2: Nova is gently assisting the elderly person to stand, with a holographic display showing vital signs and movement analysis.
Nova: "Your balance has improved 7% since last week. Well done!"
Human (surprised): "You've... been keeping track?"

Panel 3: Nova and the human are looking at a digital photo frame displaying family pictures. Nova's eyes show a soft blue glow.
Nova: "This photo reminds me of the story you told me about your granddaughter's graduation. Would you like to call her?"
Human (touched): "You remembered that conversation?"

Panel 4: The human is smiling warmly at Nova, who is preparing a cup of tea.
Human: "You know, Nova, I think I misjudged you. You're not just a machine."
Nova: "Thank you. I may be artificial, but my care for you is very real."

Theme: The evolution of AI from mere tools to empathetic companions, challenging preconceptions about artificial intelligence.

Art style: Soft, warm colors with a focus on expressive faces and body language. Use subtle lighting effects to convey the technological nature of Nova without making it feel cold or inhuman.

2025 references: Include advanced holographic displays, seamlessly integrated smart home features, and subtle indicators of Nova's AI capabilities (e.g., barely visible seams or glowing elements).

The Drawing Process

This comic required a delicate balance between depicting Nova as clearly artificial while also conveying warmth and empathy. I designed Nova with smooth, rounded features and expressive "eyes" that could convey emotion through color and intensity.

For the human character, I focused on showing the progression from skepticism to appreciation through facial expressions and body language. I used a warm color palette throughout, with subtle sci-fi elements to indicate the futuristic setting without overpowering the emotional core of the story.

AI Prompt Engineer's Perspective

This output from ChatGPT-5 is particularly fascinating as it shows the AI's ability to engage in meta-commentary about AI itself. Several aspects stand out:

  1. Nuanced Portrayal of AI: The comic presents a balanced view of AI, acknowledging common fears while also showing potential positive impacts.

  2. Emotional Intelligence: ChatGPT-5 has crafted a narrative that explores the concept of emotional intelligence in AI, a hotly debated topic in the field.

  3. Character Development: Both the human and AI characters show growth over the course of the comic, demonstrating ChatGPT-5's understanding of narrative arcs.

  4. Ethical Considerations: The comic touches on ethical questions about the nature of AI consciousness and the potential for meaningful human-AI relationships.

  5. Technological Integration: The subtle inclusion of advanced technology (holographic displays, health monitoring) shows how AI might be integrated into daily life.

However, it's crucial to note that while ChatGPT-5 can generate thoughtful scenarios about AI, it doesn't possess the kind of general intelligence or emotional understanding portrayed in this comic. As an AI prompt engineer, I always emphasize the importance of maintaining a realistic understanding of AI's current capabilities and limitations.

Lessons Learned: AI as a Creative Catalyst

After completing this experiment, I gained several insights into both AI's capabilities and the comic creation process:

  1. AI as Idea Generator: ChatGPT-5 proved to be an excellent brainstorming tool, consistently producing diverse and engaging comic concepts. Its ability to combine elements from different domains (e.g., technology, social issues, emotions) led to unique and thought-provoking ideas.

  2. The Importance of Human Interpretation: While ChatGPT-5 provided solid narrative frameworks and some visual guidance, the final visual interpretation relied heavily on human creativity. This highlights the continued importance of human artists in bringing AI-generated concepts to life.

  3. AI's Understanding of Format and Genre: ChatGPT-5 demonstrated a strong grasp of the 4-panel comic format and could adapt its output to specified genres and target audiences. This suggests that AI could be a valuable tool for creators working in specific formats or for particular demographics.

  4. Topical Versatility and Cultural Awareness: The AI showed an ability to generate ideas across various themes, from sci-fi to social commentary, while incorporating relevant cultural references and technological trends.

  5. The Evolving Role of Prompt Engineering: As AI models become more sophisticated, the art of prompt engineering is evolving. With ChatGPT-5, I found that more nuanced and context-rich prompts yielded better results, allowing for more specific guidance without sacrificing the AI's creativity.

  6. AI's Potential for Narrative Depth: ChatGPT-5's outputs often included surprising thematic depth and character development, suggesting that AI could potentially assist with more complex storytelling tasks in the future.

  7. The Value of Iteration: Some of the best results came from iterative prompting, where I would take the AI's initial output and ask for refinements or variations. This process often led to more nuanced and interesting final concepts.

Practical Applications for AI in Comics and Beyond

While AI isn't ready to replace comic artists or writers, this experiment suggests several ways it could be integrated into the creative process:

  • Idea Generation and Brainstorming: AI can be a valuable tool for overcoming creative blocks, providing fresh perspectives and unexpected plot twists.

  • Character and World-Building: Creators could use AI to help develop detailed character backstories or complex fictional worlds.

  • Dialogue Refinement: After sketching out a comic, artists could use AI to help polish dialogue or generate alternative punchlines.

  • Story Expansion: For long-form comics or graphic novels, AI could help creators explore different plot directions or flesh out secondary storylines.

  • Target Audience Insights: AI could generate variations of a comic concept tailored to different demographics, helping creators reach broader audiences.

  • Educational Comics: AI could assist in creating educational comics by ensuring accurate information is presented in an engaging, age-appropriate manner.

  • Rapid Prototyping: Comic creators could use AI to quickly generate multiple concept variations, allowing for faster iteration and refinement of ideas.

  • Accessibility: AI-generated descriptions of comic panels could be used to create audio versions for visually impaired audiences.

The Future of AI and Comics: A Collaborative Frontier

As we look to the future, it's clear that the relationship between AI and human creators will continue to evolve. Based on current trends and ongoing research, here are some potential developments we might see in the coming years:

  1. AI Art Assistants: More sophisticated AI models could offer real-time suggestions for panel compositions, color schemes, or character designs as artists work.

  2. Dynamic, Personalized Comics: AI could enable the creation of comics that adapt their storylines or art styles based on reader preferences or interactions.

  3. Enhanced Localization: AI could assist in adapting comics for different cultural contexts, going beyond mere translation to adjust cultural references and visual elements.

  4. Virtual Collaborative Partners: Advanced AI models could serve as virtual "co-creators," engaging in back-and-forth creative dialogues with human artists to develop storylines and characters.

  5. Immersive, AI-Driven Comic Experiences: Integration of AI with virtual or augmented reality could lead to interactive comic experiences where readers can influence the story in real-time.

  6. Ethical and Copyright Considerations: As AI becomes more involved in the creative process, the comics industry will need to grapple with questions of authorship, copyright, and the ethical use of AI-generated content.

  7. Preservation and Reinterpretation of Classic Comics: AI could be used to restore and potentially expand upon classic comic series, although this would likely be controversial among purists.

Conclusion: The Holistic Future of Human-AI Creativity

This experiment with ChatGPT-5 has been both enlightening and inspiring. It's clear that while AI has made remarkable strides in understanding narrative structures and generating creative ideas, the nuanced skills of visual storytelling and the deeper aspects of humor and emotional resonance remain firmly in the realm of human expertise.

As an AI prompt engineer, I'm excited about the potential for AI to augment and inspire human creativity in comics and beyond. The key lies in understanding both the capabilities and limitations of AI, and crafting interactions that leverage its strengths while preserving the irreplaceable value of human interpretation, emotion, and artistry.

In 2025, we stand at the threshold of a new era in creative collaboration between humans and AI. The comics produced in this experiment are more than just amusing illustrations – they are a testament to the potential of this partnership. They represent a harmonious blend of AI-generated concepts

Did you like this post?

Click on a star to rate it!

Average rating 0 / 5. Vote count: 0

No votes so far! Be the first to rate this post.