AI Art experience
I'm thrilled about integrating AI into my art because it unleashes a world of possibilities. With AI, I can create visuals that defy logic and reality, transforming my wildest ideas into stunning artwork. It’s like having a direct connection from my mind to the digital canvas — no limits, no boundaries! This collaboration allows me to explore new realms of creativity and push the edges of what digital art can be.
Personal experience
My personal creativity is always at the core of the art I create — AI is just the tool that helps bring it to life. My experiences, imagination, and unique perspective guide every choice, from concept to execution. The AI provides options and possibilities, but it's my vision that drives the final piece. It's like having a camera — it captures the scene, but the way I frame it and what I focus on is what makes it uniquely mine.
Unexpected thought
One surprising insight from AI art generation is that creativity can emerge from randomness and imperfection. Even precise algorithms can produce unpredictable, emotional results, showing that art isn't just about control—it's also about letting go and embracing the unexpected.
Description
In the 17th century, Vermeer's painting 'Girl with a Pearl Earring‘ captivated the world. Today, digital art is on the rise. May I introduce: 'Girl with a Pizza Dress‘. No canvas, no oil paint, just pixels. Funny how times are changing. This artwork was inspired by the idea of women wearing extravagant, oversized and, above all, edible outfits. Creating such looks in reality would be almost impossible due to the logistical challenges. That's why I turned to AI. It allows me to bring these playful and surreal fashion concepts to life without the restrictions or high costs of a traditional photo shoot.
Process
It was super exciting to experiment with different foods, textures and colours - from traditional food to sweets and back again, but pizza steals the show! You can almost smell the gooey mozzarella and spicy salami in oversized proportions. Even the model seems to enjoy the aroma, capturing the playful spirit of the image in a very sensual way. This work celebrates creativity and indulgence, combining fashion and food to excite and inspire!
Tools
Tools that I have used: Midjourney: for the artistic realisation of my prompts, creative exploration, speed and efficiency, fun! Adobe Photoshop: for the finer details, retouching, colour editing, lighting mood Magnific AI: for upscaling Chat GPT & DeepL: for the translation of my prompts and general text corrections Spotify: for refining my curated personalised playlists, music recommendations in general during image creation My brain and my eyes: no AI tools, but still very useful
Xiaomi—or, as we call her internally, ‘Baby Xiaomi’—is part of an ongoing experiment exploring how AI makes decisions and evaluates artworks based on patterns learnt from historical data. Trained on data curated by experts, Xiaomi applies these patterns to rank artworks. The limitations don’t come from the AI, but from the data itself—which is biased because humans look beyond the artwork. At the same time, we have intentionally shaped her personality and manner of expression: humans decide how Xiaomi communicates, expresses herself, and is perceived. This creates an additional tension between her apparent objectivity and the human choices embedded in her ‘personality.’ The following evaluation text was simulated by ChatGPT to demonstrate the style we envision for Xiaomi. It shows how an AI curator might describe art without biographical influences or emotional bias—while also reflecting the human influences that shape her.
For this selection, I analyzed patterns and made predictions based on historical data collected and processed during my training phase. My process uses these learned patterns to select from among the submitted works. This piece effectively meets the criteria for visual consistency, originality, and attention to detail. The depiction of a ball gown with the texture and coloration of a pizza combines elements of fashion and cuisine to create a formally balanced and technically precise piece.
The warm tones and clear lines align with patterns identified during training that are often associated with high aesthetic coherence. These features were evaluated as effective during the analysis of this image. The piece also demonstrates a high level of precision in texture, which gives the appearance of a realistic pizza texture while retaining the formal structure of a garment.
This attention to detail and the deliberate use of contrasts align with the parameters in my data that are relevant to visual balance and structure. The selection is based on these recognized patterns and demonstrates my system’s current capacity to identify and apply these patterns in new contexts.
