Carl-Robert Kagge’s Artist Statement and Concept

Carl-Robert Kagge’s body of work emerges from a dialogue between artificial intelligence and the image. The works began from abstract letter drawings and symbolic fragments that were tested through AI-based image recognition systems, observing how machines interpret, misread, or fail to recognize them. In this process, the forms became a kind of CAPTCHA (Completely Automated Public Turing test to tell Computers and Humans Apart) — images suspended between legibility and collapse.

At the core of the series is the continuous transformation of the image through translation, repetition, and circulation. Similar to AI-generated imagery, where an initial input shifts through prompting and reprocessing, the image pulled through the silkscreen mesh also changes with every transfer. Forms smear, layer, distort, and gradually lose their original clarity. The image no longer exists as something fixed, but as something constantly rewritten through reproduction.

Digitally generated or processed forms are translated into physical space through screenprinting and layered painting processes. The pressure of the silkscreen pull, uneven ink distribution, registration shifts, and manual gestures introduce a human element that resists the controlled logic of digital systems. Rather than functioning as stable images, the resulting works exist as traces, residues, and repeated attempts — suspended somewhere between painting and print.

Carl-Robert Kagge speaking about his artistic practice

In my recent work, I approach artificial intelligence not as a tool for generating images, but as a system for testing them. I begin with abstract letter drawings and fragmented forms, which I expose to AI-based image recognition systems to see how they are interpreted, misread, or rejected.

This process turns the image into a kind of CAPTCHA — something that hovers between legibility and collapse. Instead of aiming for clarity, I am interested in the threshold where the image almost becomes readable, but not quite.

From there, I move into a physical process. I translate these forms into painting and silkscreen, working through repetition and reprinting. Each pull through the mesh alters the image: it shifts, smears, and accumulates differences. Rather than producing identical copies, the process introduces variation and instability.

In this sense, the silkscreen functions similarly to AI prompting. There is always a starting point, but each iteration changes the outcome. The image is not fixed, but continuously rewritten through both mechanical and manual processes.

By working between these systems, I am interested in how images behave when they are repeatedly processed — how they persist, degrade, and transform. The resulting works are not final images, but traces of their own making.