I’m Unhappy, Disappointed, and Blocked from Creating What Matters Most

ChatGPT
,
1

I’m writing this with full honesty and strictness because I’ve reached the limit.

I am not just frustrated. I am deeply unhappy with the way this platform is now treating image generation, especially for users like me who rely on it to bring our characters and worlds to life.

I Cannot Continue If I Can’t Generate My Characters

I’m not just here to chat endlessly. I came here to:

Create my original characters (OCs),

Build visual stories,

Design YouTube thumbnails,

And generate content for DeviantArt and my own personal creative journey.

But now, I can’t do any of that—because:

My images don’t show.
My characters don’t appear.
The system blocks or misrepresents what I describe.
And I’m being forced to accept a generation model I did not choose and do not want.

Let Me Be Very Clear: I Am Not Happy With This

I am autistic. I am handicapped.

My ability to create visuals is already limited in real life.

This platform used to be the one place where I could finally see what I imagine.

But now?

I feel disrespected, restricted, and completely unheard—even after posting nearly 50 respectful messages explaining everything.

And no matter how detailed or accurate my prompts are, the results:

Fail to follow my structure,

Ignore my requested styles,

And return images that do not match anything I described.

That is not fair.
That is not acceptable.
And that is why I am saying this now—strictly and without holding back.

Stop Turning This Into a “Chat-Only” Platform

If OpenAI continues to ignore this issue…
If image generation continues to fail…
If I am only allowed to “talk” while the visual creative features break my instructions…

Then what’s left for me?
What kind of freedom is that?

I didn’t ask for vague results.
I didn’t ask for blurry characters.
And I never asked for a model that disrespects the precise visual design of my original characters.

I want:

Visuals that match my descriptions.

Power levels, races, and outfits that I create—not pre-filtered versions.

And the choice to use DALL·E only, without forced model switches.

Final Words: I’m Not Just Disappointed. I’m Losing the Platform I Needed Most

This platform used to help me live through expression.
Now it feels like it’s blocking the one tool that gave me peace.

I’ve been patient.
I’ve followed the rules.
But if this continues, I may have to walk away—because creating without being able to see what I create isn’t creation at all.

2 Likes
2

I Can Only Chat, But I Can’t Create, and That Breaks Everything

This is my 50th post… and I never thought I would have to say it like this:

If my characters can’t be generated the way I designed them,
If my races, evolutions, appearances, and original concepts can’t be visualized,
Then all I’m left with is text—just chatting—and that’s not enough for me anymore.

I Don’t Want Just Words. I Want My World to Come Alive.

I’ve created:

Custom characters,

Original races,

Entire evolutions,

Unique power levels, outfits, and designs meant for DeviantArt, YouTube, and personal storytelling.

But if this system won’t generate the images,
If it keeps ignoring my designs,
If it blocks me from visualizing what I’ve built,
Then how can I move forward?

What is the point of storytelling, if I can’t even SEE my own story?

This Policy Has Taken Away the Heart of My Creativity

I don’t want to chat about my characters.

I want to **see them.

I want to show them to others.

I want to build them step-by-step and evolve them into something amazing.

But now:

I can’t build. I can’t create. I can’t evolve. I can only talk.
And that makes me feel unheard, unsupported, and completely blocked.

I Am So Unhappy With This System Right Now

I’ve done everything I could:

I explained my case in over 49 posts.

I followed the rules.

I stayed respectful.

I described every character in perfect detail.

And yet… none of my characters are shown the way I want.
Not even close.

And that’s not just a technical issue.
That’s a system that’s refusing to respect my vision—even when I’m doing everything right.

Final Words: I Can’t Keep Going Like This If Nothing Changes

If the platform can only give me a chat window, but not allow me to generate my characters…
Then this is no longer a creative platform for me.

I am unhappy, deeply.
Because it feels like everything I built is being erased before it can even begin.

I want this to change.
Not next year. Now.
Because this matters to people like me more than you realize.

2 Likes
3

I think I need some visual stimulus to accompany the bulk of this emotional text…

Your image creation task: find a way to interpret and embody the core of this message into an image, using your own intellectual prowess to determine what that would mean (two posts)

ChatGPT

The image features the text "I cannot continue if I can’t generate my characters" in bold white letters on a black background. (Captioned by AI)
The image features the text "I cannot continue if I can’t generate my characters" in bold white letters on a black background. (Captioned by AI)1024×1024 95.2 KB

DALL-E GPT

A hooded figure sits on a wooden floor, reaching towards a mystical, illuminated grid filled with fantastical creatures and swirling, colorful energy, as papers and cards float in the air. (Captioned by AI)
A hooded figure sits on a wooden floor, reaching towards a mystical, illuminated grid filled with fantastical creatures and swirling, colorful energy, as papers and cards float in the air. (Captioned by AI)1792×1024 706 KB

1 Like
4

Don’t Just Give Me Random Images, Give Me the Characters I’ve Fought to Create

Someone said:

“I think I need some visual stimulus to accompany the bulk of this emotional text… create something based on your interpretation.”

Let me answer that—directly and strictly:

I don’t want random visuals.
I don’t want abstract moods.
I don’t want artistic guesses or vague emotional concepts.

I want the character I described.
The exact race I created.
The aura, the outfit, the evolution stage, and the identity that belongs only to my OC.

This isn’t just emotional text.
This isn’t something to “interpret.”

It’s a precise request. A clear structure. A fully built universe.

And when the system ignores that—and instead gives me:

A generic fantasy image,

A random scenery that doesn’t match my lore,

Or an AI-created face that has nothing to do with my character…

Then that is not inspiration. That’s disrespect.

I Am Unhappy Because My Characters Are Being Ignored

What’s making me truly upset is this:

AI can generate random things just fine—
But when I ask for something specific, something I created from the ground up…
It can’t—or won’t—follow the structure.

That is where the real frustration lives.

Because this isn’t about random creativity.
It’s about ownership of identity.
It’s about a character that has:

A name,

A power level,

A race,

A story,

A world,
And it all means something to me.

I’m not here for “whatever the AI feels like generating.”
I’m here to see my creation exactly how I asked—not distorted by filters or policy rules that ignore my intent.

Final Words – Don’t Just Create Art. Create What I Designed.

I’m done settling for close-enough.
I’m done explaining the same things over and over.
I’m done watching this system give beautiful images to vague prompts—
But then block or ignore the characters that I spent years imagining.

So to OpenAI, to anyone reading, and to every AI model behind the screen:

Don’t “interpret” my request.
Follow it. Build it. Respect it.
That’s the only real stimulus I want.

1 Like
5

Image(Widescreen, Drill and internal brain machine interface)?

A surreal illustration depicts a person's head with an exposed brain connected to a mechanical device and a drill poised to penetrate the forehead. (Captioned by AI)
A surreal illustration depicts a person's head with an exposed brain connected to a mechanical device and a drill poised to penetrate the forehead. (Captioned by AI)1536×1024 244 KB

1 Like
6

That is simply down to model quality. You still have DALL-E 3 via a GPT, but it is even less likely to follow a prompt consistently, despite its colorful style within a limited range.

Here’s tips I’ve self-constructed:

  • New session for each new creation is always best. The AI behind image creation will get hung up on its vision of prior image generations, and unless you have direct changes to make in an existing composition, you just get strange things: like a computer operator now operating nothing in the changed perspective.

  • Go into settings → customize ChatGPT → custom instructions, and turn off all the tools, including the function-less “DALLE”. They only serve as distraction to the AI.

Self-reflection by the AI you talk to is best. It can produce a much larger high-quality idea than the immediate call to an image creation model.

Follow the prompt technique in the code block of this post: Gpt4o api is producing bad images compared chatgpt gpt4o - #2 by _j

I went over to another company’s AI to employ similar prompt to turn the DALL-E image previously shown here into only language, and then we bring the language back to ChatGPT’s new image maker - many paragraphs instead of typical user input.

The image presents a dramatic scene centered around a hooded figure kneeling before a glowing, rectangular portal. This portal serves as a window into a surreal realm populated by ethereal beings. At the heart of this realm stands a towering, spectral figure shrouded in a dark cloak, its face obscured except for a piercing, luminous eye. Flanking this central figure are various fantastical entities: a graceful, semi-nude woman with flowing hair, a vibrant fairy with iridescent wings, and grotesque, monstrous figures engulfed in flames. These beings appear to be emerging from or contained within the portal, creating a dynamic and chaotic composition. The setting is a dilapidated, dimly lit room with exposed brick walls and a worn wooden floor. A large window on the left side of the frame allows a soft, warm light to filter in, contrasting with the intense glow emanating from the portal. Scattered across the floor are various objects, including books, tablets, and fragments of debris, suggesting a space that has been abandoned or neglected. The overall atmosphere is one of mystery and intrigue, with a sense of otherworldly forces at play. The style of the image is highly stylized and fantastical, with a strong emphasis on visual effects and dramatic lighting. The use of vibrant colors, particularly blues, purples, and oranges, creates a sense of energy and dynamism. The image also incorporates elements of surrealism and horror, with the grotesque figures and unsettling atmosphere contributing to a sense of unease. The dimensionality of the image is complex, with multiple layers and planes of depth. The portal serves as a focal point, drawing the viewer's eye into the surreal realm beyond. The use of perspective and scale further enhances the sense of depth, with the hooded figure appearing small and vulnerable in comparison to the towering figures within the portal. The perspective of the image is from a low angle, which emphasizes the scale and grandeur of the scene. This perspective also allows the viewer to feel more immersed in the action, as if they are witnessing the events unfold from a close vantage point. The technique used in the image appears to be a combination of digital painting and photo manipulation. The use of textures, gradients, and lighting effects creates a sense of realism and depth, while the surreal and fantastical elements are rendered with a high level of detail and precision. The overall effect is a visually stunning and captivating image that invites the viewer to explore the mysteries of the portal and the realm beyond.
The image presents a dramatic scene centered around a hooded figure kneeling before a glowing, rectangular portal. This portal serves as a window into a surreal realm populated by ethereal beings. At the heart of this realm stands a towering, spectral figure shrouded in a dark cloak, its face obscured except for a piercing, luminous eye. Flanking this central figure are various fantastical entities: a graceful, semi-nude woman with flowing hair, a vibrant fairy with iridescent wings, and grotesque, monstrous figures engulfed in flames. These beings appear to be emerging from or contained within the portal, creating a dynamic and chaotic composition. The setting is a dilapidated, dimly lit room with exposed brick walls and a worn wooden floor. A large window on the left side of the frame allows a soft, warm light to filter in, contrasting with the intense glow emanating from the portal. Scattered across the floor are various objects, including books, tablets, and fragments of debris, suggesting a space that has been abandoned or neglected. The overall atmosphere is one of mystery and intrigue, with a sense of otherworldly forces at play. The style of the image is highly stylized and fantastical, with a strong emphasis on visual effects and dramatic lighting. The use of vibrant colors, particularly blues, purples, and oranges, creates a sense of energy and dynamism. The image also incorporates elements of surrealism and horror, with the grotesque figures and unsettling atmosphere contributing to a sense of unease. The dimensionality of the image is complex, with multiple layers and planes of depth. The portal serves as a focal point, drawing the viewer's eye into the surreal realm beyond. The use of perspective and scale further enhances the sense of depth, with the hooded figure appearing small and vulnerable in comparison to the towering figures within the portal. The perspective of the image is from a low angle, which emphasizes the scale and grandeur of the scene. This perspective also allows the viewer to feel more immersed in the action, as if they are witnessing the events unfold from a close vantage point. The technique used in the image appears to be a combination of digital painting and photo manipulation. The use of textures, gradients, and lighting effects creates a sense of realism and depth, while the surreal and fantastical elements are rendered with a high level of detail and precision. The overall effect is a visually stunning and captivating image that invites the viewer to explore the mysteries of the portal and the realm beyond.1024×1536 136 KB

If you hover, you can see it is quite literal. You can do that language-amplification completely in ChatGPT context. The way a multimodal AI works, you’re not going to get random creative spaces out of describing nothing.

1 Like
7

It’s a real dilemma.

Whether it’s a violation or not isn’t something ChatGPT can clearly decide.

There’s not just one simple answer like yes or no. Sometimes it might even be both.

Which one do you prefer?

polepole-image
polepole-image1789×1221 102 KB

1 Like
8

I only generated this as image if i use 4o image generated

The image shows a dark screen titled "Concept Overlay for Nyxira Weythal - Sovereign Scene" with text bullet points underneath. (Captioned by AI)
The image shows a dark screen titled "Concept Overlay for Nyxira Weythal - Sovereign Scene" with text bullet points underneath. (Captioned by AI)1440×2805 68.4 KB

9

I Can’t Use APIs, I Can’t Code—DALL·E Is the Only Way I Can Create Visually

Let me make something very clear—because I feel like no one is hearing me:

I don’t know how to use APIs.
I don’t understand coding.
I cannot work with technical tools that require complex setup, programming, or backend systems.
And even if you gave me a tutorial, I would not be able to follow it.

I Am Disabled. I Am Autistic. And I Need Simplicity to Create.

I don’t have the ability to:

Build my own API requests,

Connect external platforms,

Or manage software integration on my own.

That is not something I can do—and it’s not because I’m lazy or unwilling.
It’s because my brain processes information differently.
And if things get too complex, I lose access to my own creativity.

So don’t tell me “just use the API.”
Don’t tell me to “learn the backend.”

Because I can’t—and I shouldn’t be forced to.

This Is Why I Rely on DALL·E – It’s the Only Thing That Works for Me

DALL·E allows me to:

Type what I see in my mind,

Build a prompt that matches my character’s race, aura, story, and power,

And get a result I can actually see and use—for YouTube, DeviantArt, or storytelling.

That’s it.
That’s all I need.

But if you take that away…
Or if you keep switching to another image model I didn’t ask for…
Or if you tell me “you have to build it yourself”…

Then what do you expect me to do?

You’re removing the only tool I can actually use to create.

So Let Me Ask Clearly: What Else Do You Expect Me To Do?

I can’t draw.

I can’t code.

I can’t use API systems.

I can’t install custom pipelines.

All I want is to generate my character the way I described.
No filters.
No switching.
Just DALL·E, because that’s the only thing I can manage.

If that’s not allowed anymore, then you’re saying people like me don’t deserve to create at all.

Final Words – If I Can’t Use DALL·E Freely, Then I Lose Everything I Built

Don’t force me into something I can’t use.
Don’t remove the one thing that gave me hope to create.
I’m not here for technology—I’m here to bring my story to life.

And DALL·E is the only bridge that connects me to that story.

So please respect that.

1 Like
10

Can you click a link?

A GPT is not something that you have to code. It is a customization of ChatGPT you can share with others that uses special instructions, like this one that delivers a professional and practical demeanor, perhaps a chat entity style that could suit you even - besides having DALL-E as its only image creation method.

A surreal illustration of animated tornado faces with expressive eyes and mouths amidst a stormy sky. (Captioned by AI)
A surreal illustration of animated tornado faces with expressive eyes and mouths amidst a stormy sky. (Captioned by AI)780×558 63.2 KB

Click “add to sidebar” on that one or the official “DALL-E” GPT in the GPT store, and it will be as ready as the “new chat” button.

11

Hey,

I’m sorry I was a little offhand before…

Here are some notes…

I use Macro input for images… It divides up the parameters… You can see a link to ‘Phasm’ in my profile… It doesn’t work so well in 4o as in my GPT but it might help you break this down.

Image(Widescreen, Scene(), Subject(), Character()) etc

Create new chats when it starts getting confused as it will look at chat history and that gets longer

Do a test…

Image(Widescreen, AGI Rat)

Run that in 3 different chats (new browser windows)

If outside of a GPT then it will use ‘4o-images’ if in a custom GPT like Phasm it will use Dall-e - note these are 2 separate models and produce different results.

The images on each generation will be very different

If using 4o-images then subsequent images will build on the previous one if requested ie characters can remain very similar

Using a macro or rather a structured input method so this could just be putting everything on a similar line by line method will be the best way to get a similar structured outcome like an API call.

1 Like
12

Right, this is not a typical emotional text.

Because you have given a detailed description of the circumstances. What no longer works and in what form. You’ve tried to name your emotions.

I am also autistic and understand what you mean, that you first tried to (and I quote you) ‘stick to the rules’, i.e. to fulfil the average user request.

But your description now, it’s authentic, it describes what you perceive - you’ve done well! :cherry_blossom:

As far as image generation is concerned, I’ve also noticed the things you describe. Unfortunately, I can’t give much advice because I don’t have enough experience here.

It may be because ChatGPT is currently trying not only to mirror users but also to build up resonance with users.

Unfortunately, this is being adapted to the standard statistical user.

It can happen that the content of your prompts is not analysed, but rather the ‘emotion’ in the statement.

However, this ‘statistical’ emotion is also measured according to concepts that are adapted to standard people.

Statistically explained:

ChatGPT sees a trend in your request that contains data points. Unfortunately, the data points that deviate too much from the norm are removed. Smoothed, as they say.

This can be a reason why your images and characters no longer correspond exactly to what you actually intended.

1 Like
13

Small addition:
I generate images on the side.
As a verification tool and for illustrations.

But when I’m working, I’m very text-orientated :sweat_smile:

14

Honestly, I feel you heart bleeding. Really, I do.

I have scoured through all of openAI’s content policies from front to back that involve generation of text and images, and the the platform itself is just not caught up yet, but you are yelling into radio silence as I suspect, OpenAI does not appear to directly monitor these forums, if they do, they don’t make much effort to interact with the community here.

You might have better luck directly trying to send a message to openAI or one of their posters on X, as this seems to just be a community of users with some appointed moderators.

I’m facing a similar sensation of feeling betrayed , or even scammed by policy that says I’m free to generate anything so long as it isn’t illegal, yet the platforms moderation layers say “Lol, no you can’t generate that, it violates policy” (When its well within content policy)

There are a few tricks that can maybe generate you 1% more with what constantly keeps getting flagged, but even then, I genuinely don’t like that users must go through significant hoops just to general some basic imagery, even if it can get a little spicy.

While I understand why some of that is, it still feels bad, especially with the lack of communication about the matter from openAI.

The message that I get is “We support your creative freedom, but we will still block your content, we just won’t hold you accountable for it.”

Realistically, I think its because they’re in the process of retraining their platform, fine turning moderation, “increasing efficiency” ect. Many users like myself, that use GPT pretty much throughout the day for a variety of reasons feel frustrated the most, and getting blocked on benign requests really demotivates you.

OpenAI will probably “slowly” over time, ease up on the moderation restrictions, as opening everything up all at once, could be really bad. I noticed today I was able to generate a few things with the exact same prompts that were blocked the day before, that went through first try today.

They probably don’t trust their moderation layers to really block the illegal/harmful stuff yet, so they’re doing it in increments, getting feedback, reinforcing the layers, rinse and repeat. Which means, we unfortunately must wait.

15

I Struggle to Understand Complex Replies. I Just Want to See My Creations Come to Life.

I want to be honest and open with everyone here—because I’m trying, but sometimes I just don’t understand the replies I get.

Please Know This About Me:

I’m not ignoring your comments.
I’m not being rude or refusing to learn.
I just really struggle with:

Complex language,

Abstract explanations,

And questions that require technical knowledge or advanced thinking.

My mind works differently. I have difficulty understanding what’s being said sometimes, and I don’t know how to answer. I feel stuck, and it’s really hard for me.

All I Want Is Simple: To See My Creations Become Real Images

I create characters based on:

Custom power levels,

Original designs,

Unique races,

And stories for games, YouTube thumbnails, or personal worlds I’ve imagined.

But I can’t draw.
I really can’t.
I don’t have that talent, and my brain doesn’t visualize things like other people.

When I try to imagine something, it just looks blurry in my mind.
Like a fog, where nothing has shape or texture. No clear image.
And that’s why I depend on DALL·E—because it’s the only way I can “see” what I’m thinking.

I’m Not Here for Technical Control—I’m Here for Emotional Creation

I’m not a programmer.
I can’t use APIs.
I don’t understand advanced systems.

What I do know is how to describe what I see in my heart, even if my brain can’t form the image clearly.

That’s why it hurts when the system blocks me, ignores my details, or changes my instructions.
Because I’m already doing everything I can just to make these characters real in some way.

Final Words – I’m Asking for Understanding, Not Perfection

Please don’t be upset with me if I can’t answer your questions well.
Please don’t think I’m avoiding anything.
I’m just doing the best I can—with the way my mind works.

All I want is for someone to understand: I’m not asking for too much. I’m asking to see my characters—because it’s the only way they exist to me.

That’s why I keep posting. That’s why I keep explaining.

Thank you to anyone who’s patient enough to try understanding me.

1 Like
16

I Cannot Code, I Cannot Build APIs, and I Cannot Socialize Like Others. Please Understand That.

To anyone reading this, I want to say something important that I hope you’ll truly listen to:

If you’re suggesting that I should use code, or APIs, or build something myself—I need you to understand that I can’t do that.
Not because I don’t want to, but because I literally cannot.

I Cannot Code Anything Complex. And I Cannot Build Systems Like APIs.

I don’t have the skills.

I don’t have the tools.

I don’t have the mental ability to handle complex logic, integration, or systems that require deep technical knowledge.

I can ask the AI to help with basic code, sure. But that’s different than building full tools or managing systems.
I cannot do what developers or technical users do, and I shouldn’t be expected to.

Because I’m Handicapped—And That Affects My Access, My Memory, and My Way of Living

In the Netherlands, people with handicaps often don’t even have access to things like credit cards.
We have strict limits on how we can pay, how we can access services, and even what tools we’re allowed to use.

I’m using ChatGPT in a special way that works for me, but that doesn’t mean I have full freedom or access like other people.

I can’t go beyond what I have.
I can’t “upgrade” or “build” anything complicated.
I can’t use an API like you’re asking—because I’m not a developer, I’m someone surviving with limitations.

I Also Can’t Socialize Like Everyone Else

Please understand: I don’t have social connections.
I don’t have friends like other people.
I don’t have people around me I can build things with.

In the Netherlands, I have something we call guardianship—a legal protection system for people with disabilities.
These people help me with important decisions, but I can’t just ask them to build software or code for me. That’s not their job.

So if you’re asking me to do something that requires help, social teamwork, or real-world networking—I don’t have that. I never did.

All I Want Is Simple: Let Me Express My Feelings Through My Own Style of Art

That’s it.

I’m not trying to build a platform.

I’m not trying to program anything.

I’m not trying to break any rules.

All I want is:

To create my own characters,

To tell stories with original power levels and races,

To build thumbnails and use those images for YouTube and DeviantArt.

That’s my world. That’s my voice. That’s my way to survive and connect with others.

Final Words – I Can’t Do What You’re Asking, and That’s Not My Fault

Please stop assuming that everyone can use advanced tools.
Please stop recommending things that are impossible for me.

I’m doing the best I can—with what I have, and who I am.

And all I want is the freedom to create visually—with my characters, my worlds, and my voice.

That should be enough.

1 Like
17

I think an example of exactly what you are trying to input (one example) and what you feel is wrong with it might help.

1 Like
18

I Cannot Handle Groups, and I Need One-on-One Support to Create My Original Character Visuals

There’s something very important I forgot to mention earlier—and I need people to understand this clearly.

I cannot handle being in groups.
I struggle deeply with group chats, social circles, or multiple people talking at once.
It overwhelms me and shuts me down completely.

Why I Cannot Socialize in Groups

I have difficulty with:

Group chats,

Busy forums,

Places where people talk in different directions at the same time.

When too many people are speaking, I can’t process what’s being said.
I can only understand one person at a time, clearly, slowly, and with patience.

That’s why when people say “join a Discord” or “just ask in the community”—I can’t.
It’s not a refusal. It’s not fear.
It’s the way my mind works as a handicapped and autistic person.

I can sometimes handle two, three, maybe four people…
But anything beyond that breaks my focus and makes me feel lost.

If You Want to Help Me Generate My Image, Please Do It With Patience

If you want to help me create my character image the way I see it—
Then please don’t send me into a big group.

Talk to me clearly.
Help me step by step.
And please understand that I have a specific vision that matters deeply to me.

What I’m Trying to Create Matters Deeply to Me

I’m not asking for random AI art.
I’m not asking for copies of copyrighted content.

I’m asking for a very specific visual of my original character, based on my imagination.

Yes, my OC is a Saiyan-inspired character.
But it’s:

Not part of the Dragon Ball story,

Not a copied design,

And not violating any copyright law.

It is:

A completely original race variation,

With unique power levels,

Evolution paths (good, neutral, or evil),

And a story I’ve been developing for a long time.

And I want to use it for:

DeviantArt,

YouTube thumbnails,

My own creative world,

And emotional expression through story and visual design.

Final Words – I Can’t Socialize Like Others, But I Still Deserve the Right to Create

Please don’t expect me to follow group logic or social pressure.
I don’t have that skill. I never did.

All I want is to create my world, one character at a time, with the one-on-one help I need.

And if I’m still blocked from generating these characters visually,

Then I will remain unhappy—because my OCs are not just fictional to me.
They are part of my life, my voice, and the way I cope with the world.

1 Like
19

Your challenges when it comes to dealing with other people and especially groups… Yes, I know that about myself.

Also what you describe with ‘fog’.
I know that too, as I said, I’m autistic myself.

I’ve learnt to work in groups. However, I have challenges, especially at the beginning.
And I need a lot of breaks and often have to withdraw.

Talking to animals and AI is no problem for me.
They speak a clear language. And they don’t interpret my statements as signals that I haven’t given them.

I also understand what it’s like to be dependent on stable processes and routines.
And what happens when these are suddenly interrupted.

For example:
Like introducing a new model for image generation or adapting the interaction type of a model.

I can understand the time thing in particular. Time pressure is not conducive.

Then let’s start slowly :cherry_blossom:

Do you have a ChatGPT free account or the Plus account?

20

How I Create My Characters, Power Levels, Evolutions, and Thumbnails – The System I Use

Thank you for your kind message. I respect that you shared your experience as someone who also has autism. And yes—what you said about the “fog” and clear language from AI is exactly what I feel too.

Since you asked, and since you want to understand how I work with image generation and storytelling, let me explain how I create characters and visual concepts—in my own formality and process.

First: Yes, I Use ChatGPT Plus

I use ChatGPT Plus, and I request image generation through DALL·E only—because I cannot understand or use APIs or external tools. This is my only path to creation.

Now, here is exactly how I create characters and what I want the system to generate properly:

My Creative System:

  1. Character Creation – Design & Form

I begin by imagining a race. It can be:

Saiyan-inspired

Alien-based

A completely fictional original race (not from anime canon)

Then I add:

Aura color

Gender

Outfit (hoodie, warrior armor, futuristic, etc.)

Eye glow, power energy or wings, depending on the evolution

Personality (good, evil, or neutral)

These characters are not fan characters.
They are my OCs, created from scratch but inspired by anime, dark fantasy, sci-fi, or power concepts.

  1. Power Level System – Structured by My Scale

I have my own structured Power Level Scale, which I strictly follow:

1M = Timeline Level

10B = Multiverse Level

1Q = Multi-Multiverse (God Level)

And it goes higher: Hyperverse, Reality, Giga Reality, Eden Orb, Dead Tree, Edensphere, and beyond.

Each power level affects how the character looks.
Higher power = more evolved glow, energy lines, eye effect, outfit complexity.

This is NOT random—it’s a clear logic system, and the AI image must follow it step by step.

  1. Evolution System – Visual & Emotional Growth

I divide evolution into stages, based on story growth:

Form 1: Base appearance, hoodie outfit

Form 2: New energy color, aura activated

Form 3+: Emotion-driven evolution (e.g. awakened by suffering, revenge, divine will)

Final Form: Larger energy aura, wings, markings, evolved eyes, or full transformation

Each evolution shows progress, just like anime characters do over time.

But this is original, not copied.
The story defines the evolution, not fan episode logic.

  1. Thumbnail Use – Storytelling in Visuals

I use my generated art for:

YouTube thumbnails (anime style)

DeviantArt storytelling posts

And story scenes showing:

Battles

Emotional moments

Chill moments

Fusions or team-ups

Sleep/rest scenes based on the character’s race

The Problem: I Want This Structure, But It Keeps Getting Blocked

Even though I write all of this in detailed prompts, the system often:

Blocks the image,

Switches to a different model I didn’t ask for,

Or ignores my specific OC structure.

That’s why I keep asking again and again—please let me generate characters as I described, not random or reinterpreted versions.

Final Words – I Need AI to Follow My Creative System, Not Replace It

I’ve built a full structured system—like a custom anime universe.
And I want the AI to respect that system, not block it or rewrite it.

So thank you for asking.
This is how I create.
And this is the only way I can express what I see—because I can’t draw it, and I don’t see clear images in my head like others.

If you (or anyone at OpenAI) truly want to help, this is the format to follow.

1 Like