I asked an AI to make horror character art

https://hotpot.ai/art-generator

What it does is browse the Internet for reference photographs, then uses them to create a new picture. They’re not photorealistic or anything. They are however very cool.


Doctor Loomis











Conceptually, this reminds me of Valentine with the nosebleed




This was supposed to be Jason, Michael, Freddy, Leatherface and the Tall Man playing cards. The AI decides it just wanted a bunch of Jasons.


I don’t want three Jasons, AI! I want one of each character !

Same concept here

Freddy Krueger, Jason Voorhees and Michael Myers - I think the AI used Zombie’s Michael with the paper mask on as a reference




Michael and Freddy

Michael and Leatherface

Candyman:

Tony Todd:

The AI seems to have interpreted my request for “Leatherface and Jason Voorhees” as “blend Leatherface and Jason Voorhees together”.

Wes Craven as Freddy Krueger

Kane Hodder:

Robert Englund:

Angus Scrimm, The Tall Man, Phantasm:

Phantasm Sentinel Sphere:

Hannibal Lecter:

Hannibal, Clarice, and Buffalo Bill:

The Silence Of The Lambs:

The Jigsaw killer:

They Live!





Michael and Loomis

Michael in Haddonfield

It really likes combining Jason and Leatherface.

This is supposed to be Pinhead and Leatherface.

Michael Myers vs. Leatherface

Michael Myers mask

Don’t fuck with the Chuck

“Freddy and Leatherface” gives me Freddy as Leatherface

Lament Configuration, Hellraiser

“Chucky and Jason” gives me… A Jason-style Chucky doll.

Sad Jason

Michael vs. Freddy

Darth Voorhees

]

Evil Dead Ash

Here’s its concept of Uber Jason:

Terminators, Predators, Aliens, & Miscellaneous Sci-Fi:

Here is the best it could do for a Predator:

Its xenomorphs are better:

The Terminator:

The Death Star:

Anakin:

Darth Vader:

The Emperor:

Lord Maul:

Luke Skywalker:

“R2D2 and C3P0” gave me… A golden R2.

To Boldly Go:

The USS Enterprise:

A proud Klingon warrior:

Historical and political figures. Remember, while the AI does use reference photos, the actual pictures are wholly new - it uses context clues and points of comparison to design completely new images. Which is why they sometimes look off (Hitler here, for example). It isn’t copying any specific pre-existing photos, but rather combining aspects of many.

George Washington:

Napoleon Bonaparte:

Winston Churchill:

Adolf Hitler:

Joseph Stalin:

Mahatma Gandhi:

Karl Marx:

Franklin D. Roosevelt:

John F. Kennedy:

Richard Nixon:

Ronald Reagan:

Barack Obama:

Donald Trump:

Joe Biden:

Sometimes however, if the figure is too obscure, it won’t work. This John Tyler, the 10th US President. But there are only a handful of reference photos available of Tyler:

Others:

George Orwell, author of 1984 and Animal Farm:

Michael Jackson, the Prince of Pop:

Madonna:

Marilyn Manson:

Michael Jordan - it doesn’t do well with writing.

Jackie Chan:

These are actually really cool in a surrealist way.

It does border on plagiarism sometimes though. See that writing in the lower left-hand corner of this image?


That’s an artist signature from a source training image. I assume the background wss someone else’s first, unfortunately. There’s different elements to this image - the background came from one place, and then the AI took a look at pictures of Yautja and applied a random number generator to the results.

It does, which is why I’ve seen a big push back from the art community lately. Apparently, how it works is the AI sources and aggregates imagery from around the internet to “create” an image. So it’s effectively integrating a bunch of artist’s work into a new image.

I don’t mind it because it’s how art always functions, and if artists themselves were honest they’d mostly admit it. But the intellectual property system doesn’t permit acknowledging this.

Some of this shit is obscenely cool. Tony Todd as the Candyman.