Xenoimage Dataset

A.I

2022. Project generated during the #LAB01 Medios Sintientes from Medialab-Matadero

By their initial configuration and their productive genealogy, automated digital technologies tend to reproduce the regime of visual categorisation through their predictions, suggestions and prescriptions. There is an algorithmic non-neutrality in digital technologies, which replicate through their operationalisation of a vision that is far from reality. Machine learning, AI, IoT; this entire technological corpus of sentient media is fuelled by biased images, affected by a gendered rubric that replicates algorithmic standards and, consequently, excludes all alternative subjectivity from the physical and digital spaces they inhabit.

<

The main objective of the proposal was to activate a visual future in the field of gender disruption by relying on artificial intelligence as a new sensory visualisation tool from a xenofeminist perspective. To this end, we proposed a new typology of image, the xenoimage, a category that could define those non-representational algorithmic images capable, in their performativity, of evidencing the latent biases and problems in the programming of tools for the management of human experience, and, on the other hand, serve as a tool for generating devices that escape from automated that escape from standardised automated patterns.