Zizi - Queering the Dataset: Jake Elwes

Elwes will bring his Artificial Intelligence project,  Zizi - Queering the Dataset, to the recently launched digital art space on Dover Street. The ongoing project, which started in 2019, explores the intersection of Artificial Intelligence (A.I.) and drag performance. While Drag challenges gender and delves into the concept of otherness, A.I. is often mystified as a tool which perpetuates social bias. Elwes’ evolving project intends to celebrate individuality and encourage its audiences to reflect on the biases that exist within our data driven society.

Zizi - Queering the Dataset, sets out to tackle the lack of representation and diversity in the training datasets often used by facial recognition systems. Elwes’ site specific video installation was created by disrupting these systems and re-training them with the addition of drag and gender fluid faces found online. This causes the weights inside the neural network to shift away from the normative identities it was originally trained on and into a space of queerness. Providing insight into the machine learning system, Zizi reveals what has not yet been learnt.

Elwes will present a performance piece during his residency, more information to follow on Gazelli Art House’s Instagram page, @GazelliArtHouse.