Name

_Results

close [x]

Bardou

France

Where do you locate yourself in relation to the systems you work with?

I occupy a position of friction rather than alignment. I work with artificial intelligence, but never from inside its promise. I stand slightly aside, at the edge of the system, using it as a surface of projection rather than as an authority. My work emerges from fragments — scans, memories, cinematic residues, latent images — and from the continuities that appear when these fragments are assembled through montage. I am close to the tools in a technical sense, but distant in intent: I use them to expose their blind spots, their temporal shortcuts, their tendency to smooth over history. My position is one of translation, tension, and delay.

Where are you heading, and what is pulling you there?

I am moving toward a practice where artificial intelligence becomes less a generator than a witness. What pulls me forward is the desire to slow down automated vision, to reintroduce duration, memory, and ambiguity into systems designed for efficiency and prediction. I follow the vector of montage — not as an editing technique, but as a way of thinking — allowing images from different times, places, and states of reality to collide. I resist acceleration and resolution. I am drawn toward works that remain open, that unfold over time, and that treat AI not as a future-facing oracle but as a medium haunted by the past.

How would you describe the space your practice is currently unfolding in?

My practice unfolds in a hybrid space: between cinema and computation, between urban geography and mental landscape, between recorded reality and artificial imagination. It is a space deformed by layers — 3D scans, volumetric images, latent reconstructions, archival memories — all coexisting without hierarchy. This space is neither virtual nor physical, but a zone of passage, where images hesitate, overlap, and persist. I see it as a mutable coordinate system shaped by movement, memory, and perception, where the work does not occupy space so much as it reshapes it, inviting the viewer to navigate rather than to consume.
Published in >
The AI Art Magazine, Number 3
Shadows of the Shining, AI generation, 2025.
Shadows of the Shining, AI generation, 2025.
Bardou, Shadows of the Shining, AI generation, 2025

Description

Exploring the latent space of The Shining is to approach the film as one approaches a memory. Here, artificial intelligence acts as a second consciousness: it does not reproduce — it remembers. It reconstructs The Shining from its traces, its echoes, from what has survived the narrative. The latent space becomes a mental territory, an inner Overlook where the ghosts of the image, the logic of dreams, and the mechanisms of visual memory unfold once again. The aim is to turn the exploration of the latent space into a mnemonic experience of cinema itself.

Process

What led me to this work is a long-standing desire to explore what films contain beyond their visible narrative. Cinema does not only exist in the images shown on screen, but also in the images it leaves behind — mental images, residual impressions, moments that persist in memory without being clearly defined. I am interested in these hidden layers: the images we remember without knowing exactly why, the spaces we reconstruct internally, the atmospheres that survive the story itself. The idea of latent space offered a way to navigate these invisible zones. It allows me to treat cinema not as a fixed sequence, but as a mental landscape that can be revisited, fragmented, and reassembled. This work follows an urge to explore films as memory structures rather than narratives — to move inside their latent imagery, and to experience cinema as a mnemonic process rather than a linear one.

Tools

The work combines artificial intelligence with spatial and volumetric image techniques. I use locally trained AI models, trained on my own artworks, as a way to maintain a coherent visual memory rather than relying on generic datasets. These models are combined with point clouds, Gaussian splatting, and volumetric video, allowing the images to exist in space rather than as flat representations. Together, these tools make it possible to navigate the latent space of the film as a spatial, perceptual environment — halfway between memory, reconstruction, and cinematic experience.

Image credit:
Fractured Memories
Essay by Liri Argov

In a world where machines are often tasked with reconstructing and imagining, the artist instead asks the AI to fracture and fragment the scenery, reflecting the workings of human memory.

The overlook appears not as a narrative location but as a constellation of remembered fragments, a psychological space continuously rebuilt through recollection. This construction embraces imperfection, with blurred edges and soft corners. As the AI output coalesces into art, the cinematic journey remains deliberately imperfect. From a frontal view, the memoryscape reveals the overlook as a constellation of elements, evoking the delicate quality of a hand-painted artwork and, for me, a sense of human touch.