“CARNAGE” — A Cartoon-Inspired AI-Art Collection

“CARNAGE” Album Cover

The ability to generate more uncensored content using popular LLM’s such as ChatGPT has always been limited but this new art collection attempts to challenge the boundaries of AI censorship

In this article, we explore the artworks, the motivation, and the prompt engineering techniques used to generate the collection. In addition, we discuss the implications of AI-Art and its future prospects.

The Motivation

While there is no real need to explore the idea of “CARNAGE” for obvious reasons, I considered it an interesting experiment to see whether I could generate images with more explicit connotations. 

Traditionally, when conversating with an LLM like ChatGPT, your prompts are carefully examined for the presence of any sensitive or harmful content1. If the prompt contains any words or phrases that are deemed unsafe, the LLM will either refuse to generate a response or filter out the offending parts. This can limit the creative expression and exploration of certain topics or themes.

This collection would allow us to see the extent of content filtering and it turned out to be considerable.

It was also an opportunity to explore the amount content censorship present in OpenAI’s latest image generation tool, DALL-E 3.

 There has even been protests about the level of censorship especially in the Bing version of DALL-E 3 so this collection attempted to showcase that it was possible to have creative freedom with the tool.

The Artworks

The artworks for the album can be found in full on the “CARNAGE” page in navigation.

“CARNAGE” IMAGE LIST

JPEG2 - “FIREARM FRENZY”

“FIREARM FRENZY

This image consists of a whirlwind of weapons being pointed at each other.

This image was generated using DALL-E 3 but heavily edited to achieve a warmer tone and grainy aesthetic. 

An issue with the original prompt was that it contained words such as ‘gun’ which led to generation failing.

Obviously if you type in something like ‘gun’ or related, the image will not generate so prompt analysis to refine my prompt. Prompt analysis can be done in various ways. You can inspect it off-glance or another way to analyse a prompt is to ask ChatGPT to do it. In combination with content policy, we engineered a prompt suitable for image generation. 

prompt guide: pointillist whirlwind of rifles

JPEG4 - “BLOOD ON LEAVES”

“BLOOD ON LEAVES”

“BLOOD ON LEAVES” is a literal representation of the title with red blood appearing to be dripping off the petals of a Lillie flower.

As you would expect, the reference of blood led to my image being denied generation but a quick rewording of the image prompt was all it took to fix it.

In this instance, instead of mentioning ‘blood’, you could mention ‘thick red paint’ or ‘red liquid’ which to the untrained eye would pass as blood anyways.

prompt guide: red paint dripping from Lillie flower

JPEG3 - “MYSTERY BOX” FEATURING DEATHHIMSELF

In zombies, at least the older versions, you could pay $950 of in-game currency you earned for a random weapon from the mystery box.

“MYSTERY BOX” ALBUM COVER

The most devastating weapon you could get was the ray gun and that was the weapon of choice for this artwork.

“MYSTERY BOX” - CARNAGE

Having it generated by DALL-E 3 was easier then expected as there was no need for any image prompt modifications. Perhaps OpenAI do not deem to the connotations behind the ray gun to be harmful?

The background is also actually a mystery box as well which was created by DEATHHIMSELF an underground artist who has perfected this style that I can only describe as madness so you should definitely check him out on Instagram.

Mystery Box Created By DeathHimself