Where To Sell AI Art? – Part 2. Art Marketplaces

In this newsletter, read about:

• 🕵️‍♀️ Top Art Marketplaces & AI Art

• 🗞 News and Top Reads

• 📌 AI Art Tutorial on Using Text Weights in Your Prompts

• 🎨 Featured Artist: Melissa Wiederrecht

• 🖼 AI-Assisted Artwork of the Week

• 🤓 How to Get Started with Generative AI?

🕵️‍♀️ Top Art Marketplace & AI Art

Last week, we reviewed top stock image sites and discovered that Adobe Stock is about the only one that allows uploading AI-generated images as long as you follow all their rules and recommendations. Generative AI is at its earliest stages, and it’s great that we have at least one major stock image site that is ready to support AI artists.

However, stock image sites, including Adobe Stock, have a focus on “useful” images, the ones you can use as illustrations in your blog post, news article, or business presentation. People are rarely going to stock sites for artwork.

As we have all witnessed, generative AI can assist artists in creating incredibly fascinating and stunning images. There should be a place to sell these masterpieces. So, let’s review the policies of major art marketplaces with regard to AI-generated content.

Etsy

Etsy is a marketplace focused on handmade and vintage items, but as of today, you can find a ton of AI-generated images there. It looks like the Etsy team hasn’t yet defined their position toward AI art. It’s not on a prohibited items list, but there is also no clear message that this content is allowed.

On one hand, a great part of the Etsy community is strongly against this kind of art. You can even find a petition on change.org arguing for banning AI-generated art from Etsy. On the other hand, Etsy is not a curated marketplace, and so, even if they decide to ban AI art, it would be hardly possible for them to implement this decision. The better decision might be to support transparency among sellers and request disclosure about the items created using AI.

There is still a risk that they will stop welcoming AI art on their platform because this art product is simply not handmade or vintage. Most AI artists I know, prefer to be fully transparent about their art process, and wouldn’t be willing to sell on a platform that officially banned AI-generated art, even if it’s hardly possible to actually detect it.

So keep in mind this possibility before deciding to invest your time and effort into the Etsy shop. But meanwhile, selling AI art via Etsy is a viable option.

DeviantArt

This huge online art community balances embracing AI art and protecting original artists, whose work is used to train text-to-image generators.

First of all, they have launched their own generative AI platform – DreamUp, which basically uses a vanilla Stable Diffusion model. Users can generate images with DreamUp and post them on DeviantArt. The #aiart tag will be added automatically. Furthermore, DeviantArt allows uploading AI-generated images from external providers as long as you properly tag your submissions as #aiart.

Interestingly, monetization of images generated with DreamUp is not allowed at the moment, while AI images, generated using external technologies are available on the platform for premium download. So, it looks like DeviantArt welcomes AI artists on their platform.

At the same time, the platform takes steps to protect human-generated artwork:

• In response to community criticism, they have set the “noai” flag on by default, which means that all images are set to “Tell AI Datasets They Can’t Use Your Content” by default. But artists can explicitly signal their agreement to have images scraped.

• Also, if an artist’s name was used in a prompt to DreamUp, the resulting image will have a tag with this artist’s name, indicating that the artwork was inspired by the style of that particular artist. The artists can also opt out if they don’t want their names being used in the DreamUp prompts.

• Considering that the platform got crowded with AI-generated images, the DeviantArt Team added functionality to control the amount of AI content you see on a platform.

In general, I find DeviantArt’s approach quite balanced and smart. My only concern is their choice of Stable Diffusion to power DreamUp. It was probably the easiest way, considering that Stable Diffusion is open-source. But at the same time, we know that a lot of copyrighted images were used to train this model without any authorization (as evidenced by Getty Images watermarks appearing on some of the generated images). This could be the reason why the platform doesn’t allow monetization of DreamUp-generated images.

Anyway, DeviantArt is not fighting the technology but trying to embrace it instead. That’s a great start!

Artfinder

Artfinder is a huge art marketplace with a focus on original paintings and drawings. You can also sell limited-edition prints, collages, and digital art, among many other kinds of artwork accepted on Artfinder.

What’s not welcome on Artfinder is AI-generated artwork. The list of items that you cannot sell on Artfinder is concluded with “Images created using Artificial Intelligence (AI)”.

Interestingly, this rule goes well beyond digital images generated with AI. My husband is a traditional artist, creating oil paintings on canvas, and some of his recent paintings were inspired by images that he generated with Midjourney. All these paintings were banned from Artfinder recently, and the support team responded:

Obviously, they were able to detect AI involvement in these paintings only because of the proper disclosures that my husband provided in the painting descriptions.

Considering that there is no technology yet available to detect AI-generated images with high confidence (especially, when these are oil paintings 😅), it looks like Artfinder’s approach only punishes artists who choose to be transparent about their art process, but doesn’t really erase AI art from their platform.

I like to think that “at present” in Artfinder’s response means something. Maybe they realize that this policy will need to be reviewed in the foreseeable future if they want to keep up with the industry's advances.

Artmajeur

Artmajeur is one of the largest online galleries. Similarly to Artfinder, it focuses on original paintings. However, you’ll also find sculptures, photography, and digital art on this platform.

At the moment, Artmajeur welcomes AI art in its gallery. You’ll even find a separate AI-generated image subcategory under the Digital Arts category. Most of the AI artists on the platforms sell their works as limited-edition or open-edition prints (on fine art paper, canvas, or metal). Some works are also offered via digital licensing – you choose the license you need (e.g., media, commercial), then pay and download an image. The prices on the platform vary from $30 to $2000 and even more.

While being friendly to AI art content, Artmajeur has recently updated its terms and conditions to explicitly prohibit the use of crawlers for machine learning purposes on its site. Basically, they don’t want any more AI models to be trained on the images from their gallery.

To sum up

Art marketplaces try to balance between welcoming AI art and keeping their artistic community happy. They cannot just ban AI artwork because, at the moment, there is simply no technical possibility to detect high-quality AI-generated images and guarantee their removal from the platform.

At the same time, many art marketplaces try to refrain from explicitly allowing AI art on their platforms, because a large part of the artistic community is strongly against this technology, seeing it as a threat to their profession.

So, we’ll see how it goes, but historically we know that it’s always a losing battle to fight against technology.

🗞 News and Top Reads

• Midjourney released the fifth version of its image-generating diffusion model. The latest model can generate stunningly realistic images that are often hard to distinguish from photographs. Anatomic anomalies still happen in the output images, but not so often. If you want to enjoy this new version, add “--v 5” to your prompt.

• An academic research group from the University of Chicago has released Glaze, a tool that protects art from prying AI. Glaze adds very small changes to the original artwork that are barely visible to the human eye, but AI cannot correctly learn artistic style from these altered images. Moreover, it will interpret the artwork as a different style.

• A new open-source text-to-video model has been made available on Hugging Face. Just type your prompt and get a short video clip! However, it looks like a lot of generated videos have a Shutterstock watermark written across them, implying that the model was probably trained on Shutterstock videos without authorized access.

• Just one day later, Runway revealed its multimodal AI system Gen 2 which generates novel videos from text, images, and video clips. It has many exciting features, including text-to-video generation, image-to-video generation, stylization, etc.

• And finally, Adobe has announced Firefly, its own AI Art Generator.

  • The beta version allows generating beautiful images from text prompts, but a lot more is coming, such as getting variations of different objects on your photo, creating custom vector images, and text-based video editing.

  • Firefly was trained on Adobe Stock images, openly licensed content, and public domain content where copyright has expired, which makes its output safe for commercial use.

📌 AI Art Tutorial

Become a GOD in Midjourney with Text Weights

In this tutorial, Christian Heidorn explains how to structure your prompts and how text weights allow you to compose an image just the way you want. As a bonus, enjoy much more creative and interesting results from these enhanced prompts.

🎨 Featured Artist: Melissa Wiederrecht

Melissa Wiederrecht is a Generative Artist from America, living and working in Saudi Arabia. She chose generative art as her career after earning an MS in Computer Science in 2014. Having been fascinated by code-generated art for more than 20 years, Melissa continuously pushes the boundaries of generative art as a medium, both technically and aesthetically. 

🖼 AI-Assisted Artwork of the Week

Denis Agati: “I'm terrified of lions... after dreaming about them in a scary dream. But I admire their majesty and beauty! The picture was made with the help of AI and photoshop.”

🤓 How to Get Started with AI Art?

1. DALL-E: Creating Images from Text – introduction to text-to-image generation.

2. The DALL-E 2 Prompt Book – a guidebook by OpenAI that explains how to effectively right prompts to generate images across different domains (e.g., photography, illustration, art history, 3D artwork).

3. Best Midjourney Prompts – a guide that covers the basics of Midjourney prompts (e.g., which keywords to use to create abstract art, surreal art, minimalism, etc) as well as some more advanced options (e.g., keywords related to camera lenses and filters, imitating certain artists and photographers without using their names). Finally, they provide a list of 600+ creative text prompts for image generation.

4. Stable Diffusion Prompt Book – a prompt book prepared by OpenArt. The book discusses ideal prompt format, using modifiers to change the style, format, or perspective of the image, applying ”magic words” to improve image quality, adding negative prompts, and adjusting Stable Diffusion parameters.

Share Kiki and Mozart

If you like this newsletter and know somebody who might also like it, feel free to share Kiki and Mozart. Let’s have more people learn about AI art!

If you have been forwarded this email and you like it, please subscribe below. And welcome to the world of AI art!