From Reproduction to Generativity: Updating Benjamin in the Age of AI

AI generated image

Introduction

When Walter Benjamin published The Work of Art in the Age of Mechanical Reproduction (1935/36), he captured a decisive threshold in the history of culture: the moment when the artwork ceased to exist solely as a unique, auratic object and entered the domain of reproducibility. Photography and film did not merely multiply artworks; they transformed the very conditions of art, undermining the singularity of the original. Benjamin’s account has since become canonical, a touchstone for thinking about technology, aesthetics, and politics. Yet the story did not end where Benjamin left it. In our own time, the rise of intelligent software and generative AI suggests a new epochal shift—one in which reproduction mutates into generativity: a production of “copies without originals.”

The Genealogy of Reproduction

For the Greeks, reproducibility was constrained: casting in bronze or terra cotta, stamping coins. Most artworks—paintings, sculptures—remained irreducibly unique. The late Middle Ages introduced the woodcut, making graphic art mechanically reproducible for the first time. The printing press extended this principle to literature, inaugurating a revolution that Benjamin treated as one special case of a broader logic of reproduction.

From there, technologies accelerated. Engraving and etching expanded the reproducibility of images during the Renaissance. Lithography in the early 19th century proved revolutionary: artists could draw directly on stone, producing prints with unprecedented speed and flexibility, even fast enough for daily newspapers. Mid-19th century photography surpassed lithography by removing the artist’s hand entirely. Only the eye and the lens were needed, and reproduction could now keep pace with the unfolding of real events. Film added movement and synchronized sound, creating modern mass media. Sound recording, too, converged with these advances, leading Paul Valéry—whom Benjamin cites—to prophesy a world where audiovisual content would be as available as water or electricity.

By 1900, then, reproduction had not only transformed how art was distributed but had itself become an artistic process. In photography and film, reproduction was no longer secondary to art; it became art.

Beyond Reproduction: The AI Threshold

Where Benjamin’s genealogy culminated in photography and film, our present moment introduces a qualitatively new stage. Generative AI no longer reproduces an original at all. Instead, it fabricates artifacts without origin, artworks that do not presuppose a prior existence but emerge from patterns extracted across countless data traces. If mechanical reproduction democratized access to art, algorithmic generation dissolves the very distinction between original and copy.

This mutation marks a rupture in Benjamin’s framework. Reproduction once implied a reference to an origin, even if only to undermine its aura. Generativity produces without such reference. In this sense, AI-generated images, texts, and sounds are not copies but simulations that collapse the logic of originality itself. They are instances of what might be called art without origin.

Derrida and the Arche-Writing of AI

Here Derrida’s notion of arche-écriture offers a crucial lens. In Of Grammatology, Derrida destabilizes the metaphysical hierarchy that privileges speech as presence over writing as representation. Writing, for him, is not secondary inscription but an originary process of différance—an endless play of traces without pure origin.

AI-generated art resonates with this logic. Each work is not the reproduction of an original presence but a weaving of traces, iterability without origin. In this sense, the transition from reproduction to generativity mirrors Derrida’s displacement of the metaphysics of presence. The artwork ceases to be an emanation of an origin—be it genius, inspiration, or material object—and becomes a manifestation of pure trace, of algorithmic différance.

Conclusion

Benjamin foresaw how reproduction would transform art by eroding the aura of uniqueness. Today, intelligent software carries this story further: not only does the aura dissolve, but the very logic of origin collapses. What emerges is not reproduced art but generative art, not copies of originals but works without origin. In this new horizon, Benjamin’s genealogy converges with Derrida’s critique of metaphysics. The age of generativity is the age of arche-writing, where the artwork is nothing other than the play of traces, endlessly iterable and without beginning.

References

Benjamin, W. (2008). The work of art in the age of its technological reproducibility, and other writings on media (M. W. Jennings, B. Doherty, & T. Y. Levin, Eds.). Harvard University Press.
Derrida, J. (1997). Of grammatology (G. Spivak, Trans.). Johns Hopkins University Press. (Original work published 1967)
Valéry, P. (1964). Pièces sur l’art (Vols. 1–2). Gallimard.