Theoretical context. Four thinkers who frame FACE & FICTION – AI, Identity and Illusion: Boris Groys on visibility, Lev Manovich on software and data, Joan Fontcuberta on photographic truth, Joanna Zylinska on distributed authorship.
Boris Groys – Visibility over authenticity
For the philosopher and media theorist Boris Groys, the internet is less a means of communication than a permanent exhibition space, in which people continuously design and publish themselves. He calls this self-design. Read through Groys, visibility comes to matter more than authenticity: images gain cultural weight not by depicting reality correctly but by circulating and drawing attention.
FACE & FICTION takes this literally. My own face becomes open visual material; a trained model produces countless credible yet fictional versions of the same person. The question shifts from “Is this image real?” to “What reality does this image create?” Groys also describes the artist as an organiser and selector of images rather than a maker of objects – an apt description of the AI workflow: selecting training images, writing prompts, fine-tuning, choosing and presenting results. The trained model itself becomes an archive, a condensed collection of photographic information.
Lev Manovich – The image as programmable data
Lev Manovich is one of the most influential theorists of digital culture. Where Groys looks at visibility and the public sphere, Manovich analyses the structure of digital media itself: software, data and algorithms. In The Language of New Media (2001) he describes five principles that read today like a description of generative AI:
- Numerical representation. A digital image consists of numbers.
- Modularity. Images are assembled from independent components.
- Automation. Software takes over parts of the creative process.
- Variability. A digital object exists in many possible versions – one prompt, endless variants.
- Transcoding. An image is at once a cultural object and mathematical data.
In AI Aesthetics (2018) he argues that AI does not replace the artist but changes the conditions of cultural production; creativity arises from the interplay of people, software, datasets, models and curatorial decisions. With Cultural Analytics he studies millions of images algorithmically, shifting attention from the single work to visual patterns. As he puts it in Software Takes Command (2013), software has become our interface to the world.
For FACE & FICTION, the model trained on my face is a programmable representation: identity appears as a dataset from which countless variants can be generated. Variability and transcoding explain how photographic identity moves from documentary likeness to an algorithmically producible form. Groys asks who makes images visible; Manovich asks how software makes them.
Key books: The Language of New Media (2001), Software Takes Command (2013), Instagram and Contemporary Image (2017), AI Aesthetics (2018).
Joan Fontcuberta – Why do we believe images?
The Catalan photographer and theorist Joan Fontcuberta (born 1955) has spent four decades showing how photographic truth is constructed. For him a photograph is never objective proof; its meaning comes from context – institutions, archives, media and captions. With the term post-photography he describes a world of billions of digital images that communicate more than they document.
His project Sputnik (1997) is the clearest precedent for FACE & FICTION: the invented Soviet cosmonaut Ivan Istochnikov, supposedly erased from history, wears Fontcuberta’s own face. What he did with retouching and invented archives, I do with generative AI. Read through Fontcuberta, AI is less a revolution than the logical continuation of photography’s history – it makes visible what was always true: images are constructions.
Joanna Zylinska – Creativity as a network
Joanna Zylinska is a Polish-British media philosopher, artist and Professor of Media Philosophy at King’s College London. She connects philosophy, ethics and media theory to ask how creativity, authorship and being human change in an age of algorithmic systems. For her, creativity is a relational process between people, machines and cultural infrastructures; AI is neither autonomous nor neutral, but embedded in social, political and ecological contexts.
- AI is not an autonomous artist. A model does not make art on its own initiative; it works from human-made data, algorithms and cultural contexts.
- Authorship is distributed across artists, developers, datasets, models and users – expanding the old image of the solitary genius.
- Photography was never purely objective. Technical and cultural decisions always shape it; generative AI makes this constructedness especially visible.
- Ethics belongs to aesthetics. Judging AI art includes questions of training data, ecological costs, working conditions and social impact.
- Human and machine form a network, not opposites. Creativity emerges from their relations.
For FACE & FICTION, Zylinska provides the framework for a model trained on my own photographs: authorship is shared between photographer, dataset, model and prompt – which is why every image carries its provenance.
Key books: The Perception Machine: Our Photographic Future between the Eye and AI (2023), AI Art: Machine Visions and Warped Dreams (2020), Nonhuman Photography (2017), The End of Man: A Feminist Counterapocalypse (2018), Minimal Ethics for the Anthropocene (2014).
Four perspectives, one question
Groys explains why these images matter: they exist to be seen. Manovich explains how they are made: as programmable, variable data. Fontcuberta explains why we believe them: because of the context they appear in. Zylinska explains who made them: a network of human and machine. FACE & FICTION sits where the four meet – not asking whether AI replaces the photographer, but how algorithmically produced images change what we mean by authenticity, authorship and the self.