The Desire to Live: VivSi on Yellow, Vermeer and Fashion Made with AI
Yellow Autumn is a fashion series created with artificial intelligence by VivSi, Stes and as.sunlit, published as one of the twenty editorials in MALVIE MAIN ISSUE 72, the 190-page, perfect-bound November 2026 main issue of MALVIE Magazine. All images are AI-generated. Golden light and strict elegance meet in black tailoring and dresses lit by a single colour: yellow, on the lips, the earrings, the filigree collars, the bows and the cuffs.
Order MALVIE MAIN ISSUE 72 in print:
VivSi: the interview
Your sense of beauty began with your mother, who could sew anything you sketched, and with printed fashion magazines. What from those early years still lives in the images you create today?
Grace. Harmony. The desire to create an image that makes you say, "Oh! How magnificent! I'd wear that."
In this series, autumn is not about fading but about golden light and strict elegance. How did you arrive at this idea, and why did you choose yellow as the only "live" emotion in the frame?
Yellow = the desire to live. It's like looking at the sun and receiving energy and joy. And the combination of "heavy, dense" fabrics and bright yellow is best emphasized by black.
The yellow moves across the model like a beam of light, from the lips and earrings to the collar and the cuffs. How did you plan this rhythm, and how many attempts did it take before each image felt right?
This rhythm was discovered by experts while studying the painting "Girl with a Pearl Earring" (Het meisje met de parel) by the Dutch artist Johannes Vermeer.
(research website https://www.mauritshuis.nl/meisje)
The magic of this painting is precisely due to the fact that the gaze constantly wanders from the eyes to the turn of the head, to the earring, to the smile, creating a kind of viewing geometry that is captivating. They called this a "loop of sustained attention."
And it seemed to me that in our series of images, this "loop" was formed.
You studied civil engineering and now work with websites. Do you see anything of that structured way of thinking in the way you build an image?
Structuring helps with prompting. Sometimes I don't know what I want to achieve in the end, but I like certain colors and have an idea of what quality or detail I want to bring out in that color.
All of this needs to be clarified in the prompt, building a structure that will produce the desired result.
It's possible to ask the text AI for its opinion on how best to use this color in these details and make adjustments to the prompt.
You say the same prompt looks completely different in each AI, and that you are always looking for balance. How do you choose which tool to use, and how do you know when an image has reached that balance?
I submit prompts to several AI systems at once and see which neural network understands them best.
I compare options, change backgrounds, create nuances, and set new parameters in the prompt to achieve a "wow!" moment.
Your daughter, as.sunlit, is your co author and a young graphic designer herself. How do you share the work between you, and what does each of you bring that the other does not?
I browse various websites for inspiration, looking for images that catch my eye, then adapt the prompt to what catches my eye and begin experimenting.
The prompt undergoes many changes during this process. Some images emerge in the process. Some immediately meet my visual needs, while others require further prompting to give them the desired "feel."
My daughter suggests themes and ideas for what can be done in the process, and then completes the post-processing (contrast, clarity, retouching). Although we leave some images "as is" because they lose their vivid spontaneity during processing...
But I want to reach a prompt that doesn't require further processing.
Yellow Autumn credits
Fashion Designer: VivSi
Fashion Designer: Stes
Fashion Designer: as.sunlit
Comments