Making mazes with AI: DallE2

Welcome to the 2nd in a series of posts where I will test AI image generators and see how they handle making maze art. I will be asking 10 prompts and seeing what gets generated. My goal is to evaluate different AI image sites against each other to see how they perform. In my first post I discussed the project, and today we start with the first site evaluation: DallE2.

Dall-e logo

Making Maze Art with DallE2

You can access the website here. You must sign in. As of this post there is no limit to the number of prompts you can ask for, and the site is free to use. Each prompt, or generation with DallE2 will generate 4 image options. For this exercise I chose the one closest to what I had asked for, or the most interesting. Let’s get started.

Prompt 1 - Make a medium difficulty maze of the Eiffel Tower in black and white with arrows at the start and finish

DALL·E 2  Eiffel Tower

I see the Eiffel Tower. In fact I see 2 of it for some reason. I see a maze like structure. It is black and white. But I see no arrows, and the maze is not really a maze.


Prompt 2 - Draw a medium difficulty large maze of the Empire State Building with the start and goal embedded in the structure

DALL·E 2 Empire State Building Maze

Not sure what this is. I changed the prompt vs. #1 to get rid of the arrows…I like the mazelike structure on the top of the page. And there looks to be a small Death Star under it. I think maybe DallE2 can’t make mazes.. just mazelike drawings ?

Prompt 3 - Draw a difficult maze of the White House pixel art style

DALL·E 2  White House Maze

It is pixel art style. Despite having no start and goal it looks the most mazelike of anything generated yet. And there is a house, although it is grey and looks like a dirty igloo. Interesting. Not winning any awards.

Prompt 4 - Draw a difficult maze that looks like a drawing of a famous building in sketch style

DALL·E 2  Building Maze

After asking for 3 famous buildings and seeing DallE2 go 1 for 3 I let it pick it’s own famous building and decided to try ‘sketch’ as the style. It got the style down, it looks mazelike (but not a real maze) on top of a large building with an arch entryway.

Prompt 5 - Draw a maze in the style of doyoumaze.com of a skyscraper in NYC

DALL·E 2 Maze

Nope. Maybe I’m not well known enough.

Prompt 6 - Draw a maze in the style of Sean C Jackson of a scene from a large outdoor market

DALL·E 2 Maze

So we try a famous artist of mazes. I don’t see an outdoor market but I see a lot of interesting squiggles. The maze-like shape from this perspective looks cool. Maybe there is an outdoor market there with some imagination.

Prompt 7 - Make a maze of a slice of an orange in color

DALL·E 2 Orange Slice maze

I simplified the ask and got the best technical answer yet. A start and goal would even make this a real maze !

Prompt 8 - Make a maze integrated on top of a photograph of a king sitting on his throne looking cantankerous beside his beautiful queen

DALL·E 2  King Maze

Cool 3d looking maze in the background ! The king is beside his queen…but really behind his queen. The look on his face is something more Frankenstein than anything else. And the hands are a problem (of course).

Prompt 9 - Make a solvable maze that is very large and very difficult to solve because it is so complex

DALL·E 2 hard maze

I wanted to test the maze only capabilities. Fail. What is with the red and grey items throughout ?

Prompt 10 - Make a 3d render of a red and blue glossy cube maze

DALL·E 2 3D cube mazes

4 shown because all were interesting and none was better than another. I feel like these images would be fun to walk into.

How did Dalle2 do ? Eh. It can’t make mazes, just maze looking structures. So you get some interesting outputs, sometimes, but nothing great. The 3D portion is worth exploring more !

Coming next: Making mazes with AI: Stable Diffusion

An exploration into AI Image Maze Generation

AI generated art is changing and challenging the art world. There are questions we are dealing with that are coming at us quickly and have yet to be answered by society (or in some cases, the courts). What is ethical and unethical when creating AI artwork ? Can you copyright AI artwork ? (So far courts have said they will rule on a case by case basis). What learning inputs can and should be used to create AI art (and what should not) ? There are already many lawsuits about this. Can an AI generated piece of art be entered into a contest against humans and win ethically ? Will fake images be created of newsworthy events that are fake ? Yes. In the past US courts have said they will only grant copyrights to human beings…maybe you remember the lawsuit about the selfie by the monkey Naruto. Now courts will have to deal with AI and all the layers that entails. I don’t know the answers to these questions, but they will be figured out soon (probably). What I want to explore with you is the making of a specific kind of art using AI, maze art. It is what I like to do, and what this entire website is dedicated to. In a series of posts, I want to size up the AI competition. There will not be a focus on humans with their complicated hands and fingers to mess up the AI image generators, so I think we will be ok.

I picked 10 prompts which I am going to use on a variety of AI image sites (honestly - the free ones for now) and see what they come up with. I am not sure what I will find, but I am going to take you along for the ride. I am excited to see what they come up with.

Here is an overview the AI artwork blog post series:

  1. Explanation of the exercise - An exploration into Al Image Maze Generation (this post)

  2. Making mazes with AI: Site A

  3. Making mazes with AI: Site B

  4. Making mazes with AI: etc…

  5. Making mazes with AI:

  6. Making mazes with AI:

  7. Making mazes with AI:

  8. Making mazes with AI:

  9. Making mazes with AI:

  10. Making mazes with AI:

  11. Making mazes with AI:

  12. Comparison of 10 AI generating websites - Who did each maze the best ?

  13. Learnings from making mazes with AI, look to the future

What do I hope to accomplish ?

  • See what the competition is. How good is the output.

  • Become more familiar with what AI image generators are capable of

  • Evaluate and compare the different sites to determine how well they create based on my prompts

  • Be inspired (hopefully)

  • Answer the question: Can I compete with AI generated maze art ?

For my featured image for this post I asked StarryAI (one of the AI text to image generators I will review in the series) to generate an image based on the title of this post. Here is what I got:

A picture representing AI

Not what I expected, but certainly interesting ! Hope you enjoy the series.

Maze of the Week #66 - Jallianwala Bagh Memorial Maze

Maze of the week #66 is of a memorial located in Amritsar, Punjab, India. It is the Jallianwala Bagh Memorial which sits on the location of the Jallianwala Bagh massacre in 1919. Here is a video without audio showing a walkthrough to the memorial.

Photo: flickr, P.L. Tandon

Jallianwala Bagh Memorial

Jallianwala Bagh Memorial

The Maze: A simple full color maze.

Jallianwala Bagh Memorial Maze

Jallianwala Bagh Memorial Maze

I hope you enjoy the maze. Free maze download is available on the site.

Coming next week:

The most famous building in India

Maze of the Week #65B - Bridge Guardian Maze Updated and improved !

Maze of the week #65 is of the Guardian from the bridge that overlooks Progressive Field in Cleveland, Ohio. But then I decided it could be improved…so I developed a new version of it and I will let you decide which you like best.

Photo: This photo from flickr, and Heather McLaughlin, will become important.

The Original Maze: Now called the “Before”

Bridge Guardian Maze

And now what I did to improve (hopefully) the maze and make it more interesting.

1. Colors changed - The initial maze was one color. I changed the overall color to be darker and added a secondary color. I also added shadows in many places.

2. Details added - I went back and found a better photo of the truck and added additional details to it. I added some new details to the face, added fingernails, etc. You can compare the mazes and find even more !

3. Extended the maze - I made the maze longer and extended it to the bottom of the page. The actual structure is very large and the previous “floating” version of the maze was not accurate. I also added the side panels that were missing from version 1.

4. Changed the Maze - I kept this an easy kids maze. I changed the non-core maze portion at the top and deleted the lower portion of the maze.

5. Added background - I wanted to try this as mixed media. I thought about doing a blue sky but the location of this bridge overlooks the baseball stadium, so I found the photo above to make this a mixed media maze !

6. Start and goal changed - I felt that this would look better by changing the arrows to an internal lettered start and goal.

7. Letter and branding moved - I moved the branding and lettering to the top of the page. I also re-sized the title to be larger than the city name.

And so here is the “After”. Which maze do you prefer ? Did this improve the maze ?

Guardian of Traffic Maze with stadium

I will add this maze to the current homepage and you can choose the one you want to solve ! Happy mazing !!