This impressive AI tool gives artists control back and “reimagines how AI art should work.”

When you make a purchase through links in our articles, Future and its syndication partners may earn a commission.

Key features of Blendbox

Modular Image Control allows users to customize individual elements such as texture, light, shadow, color schemes and object placement, offering incredible versatility.

Real-time adjustments ensure that users can instantly adjust these settings without having to re-create the entire image, resulting in faster iterations and fewer retries.

Multi-level output control The feature allows users to manipulate each layer of their image, similar to Photoshop, giving them detailed control over the final composition.

Blendbox is a new AI-powered painting app that moves away from the “fast jockey” approach of Midjourney and Stable Diffusion, with the goal of giving control back to artists. I’ve seen a demo of this new 2D generative AI app and it looks like next-generation photo processing software that offers control and predictability in the final image.

Adam Levine, co-founder of Blockade, told me that Blendbox’s goal is to “reimagine how AI art should work” and put creative control back in the hands of artists. This and other new AI tools are designed to move away from the “age of operational applications” and towards something more useful. “What we really wanted to get away from was the idea of ​​a monolithic prompt,” Adam reflects.

While I saw a lot of new AI tools showcased at Adobe’s MAX Sneaks that aim to give more control over AI output, including an impressive design concept, we’ll have to wait for them to arrive. Unlike Blendbox, which is out in alpha today – visit Blockade’s website to try it out.

An interesting aspect of Blendbox is that it eliminates the need for long, complex and confusing text prompts to create the final image. While text-to-image prompts are used to set the initial style and content, if you have images in a scene, you can drag and drop them to adjust the underlying artwork, change the tone of the scene, or combine images to create new results.

AI “luck” gave birth to “art” that plagued AI art applications, making them look more like slot machines than art software, and eliminating human imagination, seemingly removed in Blendbox.

In my demo, I see how Blockade CEO, artist Marguerite de Courcelles, can select styles from a drop-down menu before moving images, objects and effects around the scene to change the image and influence its composition and content. The image updates instantly and changes are saved as “steps”, so it’s easy to downgrade to an older version if you don’t like the results. I’m told that future versions will have the ability to limit “steps” by locking in styles, which suggests that Blendbox might be a good tool for implementing an idea without straying too far from the core style.

Image changes occur locally, so you do not change the entire image, but its areas and individual elements; adding a waterfall to a jungle doesn’t turn the scene into a waterfall, but rather mixes a “concept” with a scene that can then be refined and re-lit. textured and sophisticated. It also uses a layering system so you can move “concepts” in and out of the scene, influencing how they blend and develop the image.

“This is a next-generation photo editing tool,” says artist and Blockade CEO Marguerite de Courcelles. “It just allows me to think about things deeper and faster.”

Blendbox’s offering puts creativity back in the hands of artists who really want to use AI but have proven too ad hoc so far. What I learned from the demo is that those artists who understand art theory and technique—composition, form language, values, shape, texture, and color theory—will get more out of this AI app because you do have control.

“It just allows me to think about things deeper and faster,” Marguerite says. “It challenges my creativity on my part, but I think (people) will find it inherently funny.” – says Margarita.

Blendbox currently uses Stable Diffusion, but Adam explains how the app will choose from available AI engines depending on what the user wants to create. “We’re going to change engines, allowing people to choose which engine is right for them,” he says.

The mention of Stable Diffusion raises the issue of artists’ copyrights. For Adam, this is a matter of law that will be decided elsewhere and will affect the creators of the engine. “We are not in the business of making models. We are engaged in the production of products that use models. And so we fortunately see this as someone else’s problem,” says Adam, adding, “We think there will be good solutions, and we’re happy to offer these engines as well.”

The Blockade team speaks openly about the broader challenges facing AI art, but is also excited about the future and how new approaches to AI and art can be developed. Blendbox is not the team’s first AI application. The previously released Skybox AI is an application for creating 360-degree panoramic scenes in 8K resolution. Adam hints that there are plans to tie the two together, as well as ambitions to develop new dimmable lighting and 3D asset integration for Blendbox.

Having seen this in action, Blendbox definitely had the feeling that this is a release that will shift the conversation away from AI replacing artists and towards tools that artists can use and control the results. Blendbox is released today, so you can try it out for yourself.