Artistic Experience in Creating Visual Works through the AI Art Generator Midjourney

Abstract

The rapid advancement of Generative AI and Text-to-Image Generators has introduced a paradigm shift in art production, challenging traditional notions of creativity and authorship. This study aims to examine the “artistic experience” of creators in the process of producing visual works using the Midjourney platform, specifically within the context of urban photography (portrait, street, and landscape). Employing a qualitative descriptive method, this research utilizes Terry Barrett’s theory of photography criticism as a methodological bridge to translate visual elements such as Subject, Form, Medium, and Style into precise linguistic prompts. The findings reveal that while Midjourney can successfully reconstruct the “image logic” of a scene based on text, it introduces a significant “aesthetic deviation.” Comparative analysis shows that AI- generated images tend to be hyper-realistic, displaying “more mature” color tones and a cinematic polish that often removes the raw, authentic grit found in human-captured urban photography. Furthermore, the study identifies a “framing effect,” where direct comparison paradoxically heightens the appreciation for human creativity. The research concludes that the integration of AI into the creative process does not eliminate the artist but fundamentally shifts their role from a technical operator of visual machinery to a linguistic architect. In the urban creative ecosystem, this implies a new form of literacy where the ability to articulate visual concepts textually becomes as crucial as the ability to capture light, democratizing artistic creation for urban society while raising new questions about the ontology of authenticity.