Memory Recall

PREVIEW
Memory Recall
Human memory reconstructs what was experienced.
AI reconstructs what is statistically plausible.
Memory Recall is a three-channel synchronized video installation built from archival photographs of the Fondazione Giorgio Cini. Its structure is also its argument.
How memory actually works
Human memory does not store the past as images in an archive. It does not preserve events intact.
When something happens, the brain records fragments: a face, a sound, a bodily sensation, an emotion. These elements are distributed across different regions of the brain. What we call a memory is not a file, but a pattern temporarily assembled from these traces.
Attention, emotion, and context determine what leaves a mark. What matters is not everything that occurred, but what was felt, noticed, or feared.
Memory begins as a constellation, not a document. To remember is not to retrieve. It is to rebuild.
Every act of recall is a rewriting
When we recall an event, the brain reactivates fragments of past perception and stitches them together in the present moment. The image, the emotion, the bodily feeling reappear, but never in exactly the same way.
Each act of recall briefly destabilizes the memory. New emotions, new meanings, and new perspectives can enter before the memory is stored again. For this reason, memory changes every time it is remembered. Not because it is unreliable, but because we have changed. Because memory is alive.
Memory is shaped by the body as much as by the mind. Posture, tension, comfort, and avoidance often remember what words cannot.
What we carry forward is not a factual record, but a living residue: an emotional imprint, a sensation, a narrative that continues to evolve. The past is continually rewritten to make sense of the present.
The installation enacts this. The first channel holds the archival photograph as historical trace. The second is its AI-generated reanimation, filling the gaps between moments with probability, style and inference. The third is animated from Florencia S.M. Brück’s own photographic archive and diary notes, standing in for what the archive cannot store: sensation, emotion, embodied memory, the memory that moves through the body.
A voiceover, recorded with an AI model trained on the artist’s own voice, asks what survives in an image, what a machine perceives when it looks at the past, and what kind of memory each of us will leave behind. What a machine reconstructs is pattern and probability. What a human remembers is smell, weight, grief, warmth. This is not a technical limitation waiting to be solved. It is a fundamental difference in the nature of knowing.
In this work, the archive is not treated as fixed evidence. It becomes a site of transformation. What you see here is not what happened. It is what remains.
Memory Recall does not propose AI as the new memory of culture. It uses AI as a structural device to reveal what AI cannot reach.

What this work asks
What survives in an image? What does a machine perceive when it looks at the past? And can an intelligence without a body ever truly remember?
Current AI systems, however sophisticated, are structurally blind to the sensory, emotional and embodied dimensions of human memory. What a machine reconstructs from an image is pattern and probability. What a human remembers is smell, weight, grief, warmth. That missing layer is not a technical limitation waiting to be solved; it is a fundamental difference in the nature of knowing.
Rather than a linear narrative, Memory Recall unfolds as a constellation, echoing how memory itself behaves: fragmentary, recursive, coloured by emotion and time. The third channel exists precisely to name what the algorithm cannot reach: the memory that moves through the body’s genetic code, that gathers in rituals and gestures passed from hand to hand, that lives not only within us but between us. The memory that has no image, and yet endures.
Production pipeline
From source material to projection: the voice, the archive, the AI reconstruction and the sensory layer converge in post-production into a synchronised three-channel work, exported in 4K.

Narration
Voiceover written by the artist and recorded using an AI model trained on her own voice.
Here, in this monastery,
I walk through cloisters built for silence.
What was once a house of prayer now shelters the archives.
These walls hold centuries, like breath held in stone.
This place was reborn from a father’s grief,
a son’s life broken in mid-flight.
From that wound he shaped not stone for mourning,
but a living archive,
so that memory might endure where life could not.
Every place and event carries the record of its time,
written with the tools and technologies at hand.
And the way they are recorded
determines how they will be remembered.
The outcome of a war is less gunpowder than pen.
In the Refectory, where Veronese’s Wedding at Cana once stretched across the wall,
there is only its echo,
a facsimile restored to the place it was meant to live.
Because here, memory is not in the pigment,
but in the dialogue between image and place.
Memory does not preserve the past as it was.
It reshapes it each time it is recalled,
shifting with perception, with emotion, with imagination.
The past is not behind us.
It is alive within us.
It awakens when summoned,
by a glimpse, by a word,
by the mandarin fragrance you thought forgotten,
by a song that still echoes in your chest,
by the salt of the sea upon your lips.
Memory lives not only within us, but between us.
It gathers in rituals and dances,
in sites and landscapes,
in gestures passed from hand to hand.
There is no single truth.
Even death does not dissolve it.
It moves through the body’s genetic code,
a whisper of those who came before.
Memory defines the shape of our lives.
It gives the present its context,
resonant with everything we have lived,
woven into the story of who we are.
And I wonder,
what kind of memory will you leave behind?






Archival sources
Archivio Eleonora Duse, Istituto per il Teatro e il Melodramma, Fondazione Giorgio Cini.
Archivio Aurél M. Milloss, Istituto per il Teatro e il Melodramma, Fondazione Giorgio Cini.
Raccolta su schedoni, Fototeca dell’Istituto di Storia dell’Arte, Fondazione Giorgio Cini.
Special thanks
D.A.I.R. Evaluation Committee:
Chiara Casarin, Ennio Bianco, Veronika Liebl, Luca Massimo Barbero, Sara Tirelli, Valentina Venturi, Centro Digitale — ARCHiVe, Fondazione Giorgio Cini.
Marianna Zannoni, Istituto per il Teatro e il Melodramma, Fondazione Giorgio Cini.
Monica Bassanello, Fototeca dell’Istituto di Storia dell’Arte, Fondazione Giorgio Cini.
Eva Salviato, Centro Studi di Civiltà e Spiritualità Comparate, Fondazione Giorgio Cini.
Marta Zoppetti, Centro Vittore Branca.
