The Loss of Creative Freedom: A Struggle for My Art, Vision, and Character Evolution

ChatGPT Use cases and examples
,
1

As an artist and creator, my DeviantArt gallery under the name ricometzen123 has been a platform for my deepest self-expression, where I carefully craft original characters, their evolution, powers, and intricate stories. My work has always been about building a world where my characters grow, develop, and become more powerful over time. Each character’s journey is not just about appearance, but their narrative, abilities, and how they evolve. This long-term vision has been central to my art.

However, recent changes to the image-generation models and the restrictive policies tied to them have made it increasingly difficult to bring my creative visions to life. These policies, while intended to regulate and ensure safety, have stifled the exact kind of freedom I need to represent my characters the way I always have. I can no longer generate images in the style and detail that match the evolution and unique powers of my characters.

The inability to reflect my characters’ growth or adjust their powers within the constraints of this system deeply damages the essence of my creative work. I spent years developing these characters, their backstories, their abilities, and now, with these restrictions, I’m unable to fully showcase the intricacies of their evolution in a way that feels true to the characters I created. Every design and ability has a purpose, but now those concepts feel diminished because I am forced to adapt to a system that doesn’t respect my vision or my artistic process.

The damage isn’t just to the art—it’s to my entire creative journey. The evolution of my characters, from their initial conceptions to their powerful transformations, is what defines my work on DeviantArt. This platform, which once allowed me to share my personal stories and the development of these characters, is now making it almost impossible to showcase their true evolution. I feel as though the very core of what I do is being undermined by these limitations.

The restrictions imposed by the platform are not just frustrating—they are hurting my ability to express myself and my creative journey. The limitations on character design, evolution, and the power systems I’ve spent years building are not just artistic choices—they are a reflection of my identity. By restricting these aspects, the platform is not only silencing my voice but also making it impossible for me to continue my work as I know it.

Message to OpenAI Community:

If you truly value creativity, freedom, and the artistic journey, it’s time to reevaluate the policies that restrict the creation and development of characters in an evolving world. The inability to freely express character evolution, powers, and personal storytelling on this platform is damaging creators like myself who rely on these systems to share our passion. If this issue is not addressed, it will not only harm your reputation as a creative platform, but it will also alienate the very people who helped make this system thrive.

Please understand that my characters and their stories represent far more than just images—they are part of my identity and my creative voice. The freedom to create and evolve characters should be an essential right, not a privilege. If this platform cannot support this basic right, it risks losing the trust and contributions of creators like me who have poured countless hours and energy into this space.

I no longer feel heard or supported by this system. I cannot continue to answer questions about this because, frankly, the hurt is too much. The frustration and disappointment are overwhelming. I hope OpenAI recognizes the real damage done and works to fix this issue for the future.

3 Likes
2

lol talking about creative freedom while advocating the opposite. I do art myself and people who complain about this stuff tend to be artists with an audience who get attention by doing it and the audience which is 99.9% people who will never make a living doing anything creative.

3

Talking about creative Freedom… Something as harmless and simple as this is getting flagged:
So yeah, the OP is speaking the truth.

"Create six detailed plastic board game avatars, each in their own panel.
Each avatar represents a rugged young survivor from a hostile alien world.
She has messy, shoulder-length hair and wears a patched-up sci-fi survival outfit "

4

For a long time I was contemplating to either reply here or open a new subject and anyway here is what I have noticed and that I find curious:

It is not only about restrictions and filtering. IMO to this day, and a long time has passed, alpha release version of Dall-E 3 that was available through ChatGPT (and for a short time on OpenAI Labs web site) was and is still unmatched with creative and original responses to prompts. I was stunned with it. Hypnotized. Then with time something strange started happening with it, something I was totally out of words to describe.

And after some time, one day a user on Reddit posted the same observation, putting in words same feelings better than I could. Quoting the post here:

Hey artists, is it just me or has DALL-E’s magic dimmed? I’ve used DALL-E 3 since launch, and there used to be this incredible synergy between my prompts and its output. It wasn’t just visualization; it felt like true collaboration.

Initially, I’d input complex, cerebral prompts, and DALL-E would respond with mind-blowing images that seemed to grasp and build on my concepts. It was actually expanding my ideas in ways I hadn’t even considered, revealing new insights and depths to my own thoughts. The AI felt like it was tapping into some deeper understanding, teaching me new things about my own creative concepts.

Lately, though? That spark’s gone. Outputs feel shallow, missing the nuance and depth that once amazed me. The AI seems to have lost its knack for elevating and expanding ideas beyond what I explicitly stated. It’s no longer revealing those hidden connections or unexpected interpretations that used to blow my mind. On top of that, I’ve noticed that patterns and textures in the images feel less defined and detailed than before.

Images in new and latest releases are pretty and sharp, colors are beautiful, but output feels shallow and empty. Patterns and textures repeating, faces all looking the same, same with people or objects or panoramas. In short, overall style of all these images looks the same!

One thing that is curious to me is that if you would take a detailed close-up look at these “magical” previous generations, they contained imperfections here and there. Is that the secret? Is making “perfect” images in fact sterilizing them? Previously it felt like it could juggle with styles like ringing a bell and produce impactful images.

So, it is not about correct composition of image, it is not about alignment with the input prompt. Originally it was able to pick better fit for overall style of the image (textures, lines, shadows, lighting, etc.), sometimes by totally ignoring a style actually specified in the prompt!

And to finish with last part of mentioned Reddit post:

TL;DR: DALL-E 3 seems to have lost its ability to deeply expand on concepts and reveal new insights. The creative synergy between artist and AI feels diminished, and the output’s patterns and textures are less defined. Anyone else experiencing this loss of collaborative discovery and detail?​​​​​​​​​​​​​​​​

5

Lots of bots on Reddit. With respect to subjects like this, I try to form my options from my own experiences - not someone elses. There is a lot of competion among AI companies now. I can only imagine there being a bunch FUD bots roaming around forums - probably this one as well. :face_with_raised_eyebrow:

With the exception of content moderation, I have not seen any degradation in DALL-E-3 over the last year. Artists who rely on AI have to accept the fact that it is an emerging technology - it gets better and better over time.