Since our Banksy Guerrilla Marketing Lessons is our top April post, I wondered if we could use AI to “see” Banksy. Seventeen years ago, I heard cancer and my name in the same sentence. Not long after that depressing day, I wrote a list of things I’d like to accomplish. I wanted to ride a bicycle across America and did so in the summer of 2010.
I also wanted to meet the street artist Banksy because meeting one’s heroes is inspirational. I met Robert Rauschenberg at the Museum of Contemporary Art in Chicago. I got invited to tour Keith Haring’s studio not long after his death, and simply standing in New York in Keith’s studio was an inspiration. Finally, I spent a surreal morning with the artist Peter Max.
As an AI language model, I cannot create images or see people, so I cannot offer a visual description of Banksy. However, it's important to note that the true identity of Banksy, the British street artist, has been kept a closely guarded secret, with only a few people supposedly knowing who he or she is. The mystery around Banksy's identity has sparked much speculation and debate, but no concrete or verifiable information about the artist's appearance is publicly available.
That said, there have been some unconfirmed reports and alleged sightings of Banksy, suggesting the artist could be a white male in his 40s or 50s, with a casual, unassuming appearance. Some believe Banksy may use disguises or work with a team to remain anonymous. Please keep in mind that this information is purely speculative, and Banksy's true identity and appearance remain unknown.
I used ChatGPT’s description as a Dall-e-2 prompt.
create a painting of a white male in his 40s or 50s, with a casual, unassuming appearance in a hoodie Shepard Fairey style
Here are the images OpenAI’s AI-generated images Dall-e-2 created.
Now thanks to ai-generated art, I can "see" Banksy even if meeting him is unlikely. So now it's your turn. Click on the AI Banksy Art Competition below to learn more about how to enter what you think Banksy looks like.
Now it's your turn to fire up Midjourney, Bing Image Creator, Adobe Firefly, or whatever ai art generator and machine learning algorithms you use to share what you think Banksy looks like. I'm not going to write a bunch of legalese terms since the first place is a sheet of Banksy's Ukraine stamps, and we don't want to own your AI Artwork or turn entries into NFTs you are simply granting me and WTE a limited license to share your masterpieces on our website and social media.
Use whatever ai technology you want to create your works of art showing what you think Banksy looks like. Email your digital art inline, not as an attachment, since attachments from people I don't know can be scary. If you can draw, something I never mastered despite studying art at Vassar College, or use Photoshop or Illustrator instead of an AI bot, email your pieces of art, and I'll include your
Judging the AI Banksy Art Competition will depend on how many entries we receive. If three people and a dog enter, I will bet on the dog, figure out who wins and explain why. On the other hand, if we shock the art world and the contest blows up, we'll ask visitors who should win.
The contest starts Thursday, April 20th, and ends Thursday, May 4th. Email inline images or a secure link to your image to martin (at) wte.net. If attaching a file is your only opiton, please send your entry to my personal email mobriff (at) gmail.com that way I don't crash company servers. Thanks, Martin
The AI Banksy Art Competition winner, receives one of my sheets of Banksy Ukraine stamps showing a girl flipping Putin. The image came from work Banksy added to the war-torn Ukrainian streets.
Working on the AI Banksy Art CompetitionI discovered something called a Generative Adversarial Network (GAN). A GAN, I discovered, is a machine learning system invented by Ian Goodfellow. GANs consist of two simultaneously trained neural networks in a process called adversarial training.
The generator network wants to produce realistic samples (e.g., images, text, or other data) from random noise or latent variables, while the discriminator network's is used to determine whether a given sample is real (i.e., coming from the actual data distribution or dataset) or “fake” (coming from the generator). These two networks play in a zero-sum game, where the generator's objective is to create samples that can fool the discriminator, and the discriminator's objective is to identify real and fake samples.
During training, the generator becomes better at producing realistic samples, and the discriminator becomes better at distinguishing between real and fake samples. When an equilibrium is reached, where the generator produces samples that are difficult or impossible for the discriminator to distinguish from real data, training concludes.
GANs help generate realistic images, enhance image resolution, convert images between different styles, generate music, and creating synthetic data. Chances are your ai art generator was trained with a GAN.