Bizarre Love Triangle: Generative AI, Art, and Kitsch

Generative artificial intelligence (GenAI) has engrossed the mainstream culture, expanded AI’s creative user base, and catalyzed economic, legal, and aesthetic issues that stir a lively public debate. Unsurprisingly, GenAI tools proliferate kitsch in the hands of amateurs and hobbyists, but various shortcomings also induce kitsch into a more ambitious, professional artists’ production with GenAI. I explore them in this paper. Following the introductory outline of digital kitsch and AI art, I review GenAI artworks that manifest five interrelated types of kitsch-engendering expressive flaws: the superficial foregrounding or faulty circumvention of generative models’ formal signatures, the feeble critique of AI, the mimetics, and the unacknowledged poetic similarities, all marked by an overreliance on AI as a cultural signifier. I discuss the normalization of these blunders through GenAI art’s good standing within the art world and keen relationship with the AI industry, which contributes to the adulteration of AI discourse and the possible corruption of artistic literacy. In conclusion, I emphasize that recognizing different facets of artists’ uncritical embrace of techno-cultural trends, comprehending their functions, and anticipating their unintended effects is crucial for reaching relevance and responsibility in AI art.

💡 Research Summary

The paper “Bizarre Love Triangle: Generative AI, Art, and Kitsch” by Dejan Grba offers a comprehensive cultural‑aesthetic critique of how generative artificial intelligence (GenAI) is reshaping contemporary art practice and, in the process, re‑introducing kitsch into works that would otherwise be considered high‑brow. After a brief historical overview of digital kitsch—tracing its roots from 19th‑century debates through post‑modernist appropriation, Dada, pop art, and the rise of digital media—the author situates GenAI within the broader field of AI art, distinguishing between experimental, concept‑driven uses of computation and more pragmatic, market‑oriented productions.

The core of the article is the identification of five interrelated “kitsch‑engendering expressive flaws” that appear in professional GenAI artworks:

1. Derivative Exoticism – Artists foreground the statistical signatures of the models (e.g., neoclassical palettes, early photographic artifacts) without subverting or transcending them, resulting in works that are visually striking but conceptually shallow.
2. Unlearning Curveballs – Attempts to “avoid” model signatures lead to superficial stylistic tweaks rather than genuine critical re‑configuration, leaving the underlying bias untouched.
3. Shaky Critique – Works that claim to critique AI often do so in a perfunctory manner, inadvertently reinforcing AI as a cultural signifier and diluting the potency of the critique.
4. Obtrusive Figuration – The anthropomorphisation or deification of AI creates an immediate affective hook but masks the non‑human nature of the technology, turning AI into a mythic figure rather than a tool for reflection.
5. More of the Same – A tendency to reproduce visual or thematic motifs from earlier art history (or from previous AI projects) leads to redundancy, where novelty is replaced by statistical averaging of past styles.

These patterns are illustrated through a series of case studies: Jason Allen’s “Théâtre D’opéra Spatial,” Boris Eldagsen’s “Pseudomnesia: The Electrician,” Aurèce Vettier’s “Le travail des rêves,” fuse* studio’s “Oniric(a),” and others. In each example the author shows how the artist’s reliance on GenAI either amplifies the model’s learned aesthetics or results in a half‑hearted attempt to distance from them, ultimately producing works that are aesthetically appealing yet conceptually kitschy.

The discussion expands the analysis to two macro‑level forces that normalize these flaws. First, the art world’s institutional acceptance of GenAI—through exhibitions, awards, and market validation—confers legitimacy on works that would otherwise be dismissed as gimmicky. Second, the AI industry co‑opts artistic production as a marketing vehicle, using high‑profile artworks to showcase technology while simultaneously shaping public discourse about AI in a way that obscures its limitations and ethical concerns. This symbiotic relationship erodes critical literacy among both creators and audiences, leading to a “corruption of artistic literacy” and a dilution of public debate about AI’s societal impact.

In conclusion, Grba argues that recognizing the nuanced ways artists uncritically embrace techno‑cultural trends is essential for fostering responsible AI art. He calls for a more reflexive practice that interrogates model biases, foregrounds methodological transparency, and re‑positions AI from a mere aesthetic shortcut to a collaborative partner that can genuinely expand artistic inquiry. Educational programs, critical scholarship, and institutional policies must therefore evolve to support artists in navigating the ethical, aesthetic, and sociopolitical dimensions of GenAI, thereby preventing the perpetuation of kitsch and ensuring that AI‑driven creativity contributes meaningfully to cultural discourse.