Titled Due to the Image, the exhibition consists of the first series of paintings made by Alessandro Giannì in collaboration with his custom-made artificial intelligence program named Vasari (after Giorgio Vasari, 1511-1574).
Vasari was born from Alessandro Giannì's need to find an alternative to human logic to conceive his works. The AI has been programmed to emulate the compositional processes of Giannì, by tracing the artist's body of work and his extensive digital visual archive.
The creation starts with several inputs, such as keywords provided by the artist that Vasari uses to search the web for images, or a selection of images that are uploaded into the software and then reworked. The software can process an unlimited number of images that will remain in the archive and are stored in a non-relational database, using a deep learning model called convolution neural network (CNN).
In addition to the archive and given keywords by the artist, Giannì provides the AI with crucial input for sketch composition, such as image size, colour preferences, quantity of images to combine, and a variety of other possibilities that will be used to create the digital sketches that the artist will then select and paint in oil on canvas. Giannì also intervenes manually and feeds the resulting image to the AI, which modifies it again based on further inputs. In this way, two realities are contrasted: the programmed, endlessly repeating binary e-matrix, and the human one made of defects, errors and weaknesses. The exhibition layout was also conceived by uploading the exhibition’s images into Vasari’s database, which then came up with potential solutions for installation
The software was developed in 2020 by Unboolean, a creative studio that moves on the border between art and engineering.