My Top 10 AI Images of 2023 - - - - - - - - - - - #8 Joker Circa 1910

A nice feature of AI image generation is that you can suggest creating an image set in a specific period. For example, imagine if the Joker fought Batman around 1910. AI tends to do a better job with individual characters than with multiple people in one image. This is often because when describing several people in text, AI can lose focus on which person should receive specific details, leading to a tendency to morph two or more characters together.

[Sidenote: If you’re interested in what inspired the character Joker, check out the silent film “The Man Who Laughed”; a truly eerie depiction of a character that would influence comic book/movie culture for decades to come.]

Fabulous sepia toned image of the Joker back in 1910; the expression alone is priceless

Here are some samples of attempts to depict the two characters in a 1910 fighting scene. Note how Batman often appears more manic than his classic stoic expression. It's as if the Joker is wearing the Batman costume. 

Additionally, Batman's armor is too technically advanced to represent his costume accurately in that timeline. Apparently, it seems AI is pulling reference to the Batman’s costume mostly based on the more modern Batman films instead of the cloth based costumes in earlier times.

My Top 10 AI Images of 2023 - - - - - - - - - - - #7 Celebrity Not-Quite-Right-Look-A-Like

To me, nothing is more impressive than when an AI program successfully captures the likeness of a famous person. However, AI companies are now striving to limit this aspect of image generation due to the growing threat of Deep Fakes in multimedia. If left unchecked, this could lead to worsening political scenarios, increased legal actions from celebrities regarding the misuse of their likenesses, and a general rise in public deception, fueling falsehoods and conspiracies. This could potentially lead to complete anarchy in a world already susceptible to misinformation.

This image doesn’t quite look like Joey Heatherton, but it captures the style of her time.
The green smoky background, black dress, bobbed hair, and mystic pose makes this a fantastic image.

Most American-based AI sites, tend to not do great exact likenesses. They do permit likeness usage generally as a reference in prompts to generate similar-looking characters. But if referring to anyone famous in your prompt, the censorship restrictions are usually maximized. (This is not true from many foreign AI sites, especially those based in Japan.)

In this particular case, I referred to a lesser-known celebrity from the early '70s who I saw on TV all of the time on variety, talk, and game shows as a kid: Joey Heatherton. (Not only was she cute, sexy, and bubbly — she was the inspiration to Catherine O’Hara’s character, Lola Heatherton, one of my all-time favorite comedy shows, SCTV.)

Catherine O’Hara as the over-the-top-manic-laughing, Lola Heatherton, on SCTV in the late ’70s
Yes, that's Catherine O'Hara, known for Home Alone, Beetlejuice, Best of Show, and Schitt's Creek.

My Top 10 AI Images of 2023 - - - - - - - - - - - #6 Capt. Billy Buttons

I have a favorite character I created for a children's book named Captain Billy Buttons, who is a pirate and serves as my company mascot for my ARRRGGGHHH Ink illustrations on social media. When experimenting with AI, I typed his name into an AI art generator just to get an idea of who AI thought the name represented. The result was unexpected and insane. 

AI interpreted "Captain Billy Buttons" quite literally, covering the character with colorful, button-like pox, resembling a vibrant candy infection from a Willie Wonka candy factory. His expression is one of great surprise yet he remains at attention, as if he's bravely posing for this photograph despite his unusual appearance. 

This outcome was a delightful surprise, transforming my silly pirate name into something so whimsically bizarre. Love it!

My Top 10 AI Images of 2023 - - - - - - - - - - - #5 Mycol

Comics were a huge part of my childhood and a big reason why I became an artist. As a teen, I drifted away from them. But post-college, writers like Frank Miller with "Dark Knight," Neil Gaiman with "Sandman," and especially Alan Moore with "Watchmen," "Miracleman," "Swamp Thing," and more, pulled me back in.

Alan Moore's run on "Swamp Thing" is a standout in my opinion. He took a character I loved and flipped its whole history in just one issue. It was groundbreaking. It felt like Moore was shaking me, saying, "Think you know Swamp Thing? Think again!" 

A character I dreamt up decades ago, finally appears as I envisioned
Mycology: the branch of biology concerned with the study of fungi

Back then, I often contemplated the stories I would create if I were a comic book writer. Reflecting on Swamp Thing, Moore's distinctive twist opened up new avenues of creativity. This led to the conception of characters who could seamlessly integrate into that universe. Among them was Mycol – akin to Swamp Thing, but with a twist. While Swamp Thing represented the "Green" of plant life, Mycol personified the "Gray" of the fungal domain, a completely different realm. Spoiler: Mycol is the offspring of Swamp Thing and Abigail Arcane, unknowingly conceived through the sinister machinations of Anton Arcane.

This story lingered in my mind, yet it remained largely unexplored until the advent of AI image generation decades later. I am quite fond of how the Mycol image materialized. It closely aligns with my vision for the character, though it still lacks some additional fungal elements that I envision within him.

My Top 10 AI Images of 2023 - - - - - - - - - - - #4 The Hanged Man challenge

AI excels in numerous areas, but it also has its limitations. These limitations can be quite frustrating, particularly when you consider how AI can effortlessly generate incredibly detailed images within seconds. Yet, we still encounter peculiar errors and anomalies. As of this writing, some of the most common challenges include: generating coherent text messages, depicting realistic human hands, simulating archery skills, and my personal favorite, recreating "The Hanged Man" card from a traditional tarot card set.

This might be my favorite image of my Top 10. Quite a hauntingly beautiful image.
I added details of him being infected by a fungus - more on that later . . .

I stumbled upon this AI limitation in a Reddit group chat. A user was expressing their frustration with attempting to create a tarot card set using AI-generated images but faced a significant obstacle – they couldn't get AI to depict "The Hanged Man." Many of us in the group, including myself, attempted to generate the image, but no one came close to the traditional depiction of the card. (Which typically features a man hanging upside-down by a rope tied to one of his legs, the other leg bent horizontally, with the other end of the rope secured to a tree branch above.)

An example of the classic Hanged Man tarot card

Some argue this happens because the AI is censored from showing a hanging person. Others said, AI doesn't quite understand what "upside-down" means and still needs to learn about terms of placement, locations, directions, and alignment. I agree with the latter. Often you can not direct AI where you want things to be placed in a text prompt, it's very hit-or-miss.

My first attempt at The Hanged Man (left) and the flipped version (right)
The gravity-defying skull chains, lowered branches and bottom light source ruin the hanging aspect.
Even though he has two left hands, I liked this image enough to use as a reference for the top image.

Certain individuals believed that a potential solution to the problem was to create an image of a hanging man and then flip it upside-down to achieve the desired effect. However, this approach proved to be ineffective. This is primarily due to the presence of elements like the ground, trees, and tree branches in the background of this particular tarot card. Additionally, AI algorithms incorporate factors such as gravity's impact on clothing and the positioning of upper lighting sources in their image generation process. Consequently, when you flip an AI-generated image, it often still appears as if you simply flipped the original AI image.