Skip to content
To WdKA graduates platform
Ruben
Staps

‎‎‎‎‎‎‎‎Algorithm(e)

An ongoing research into online identity in the age of biometric surveillance technology 📹

Project details

Programme
graphic-design
Practices
autonomous-practice
Minor
Digital Craft

We use our face on digital media platforms in order to represent ourselves in the online world, a world often described as the cloud, a name that represents something that we can actually conceive, while in reality the internet can not just be seen by looking at the sky. Cables placed deep in the ocean and data-centers hidden in big boxes hide the control over its users both in online as well as the offline world. We often fail to understand the consequences and costs of being an online entity. Living online means to be seen, and to be seen is to be recognized, to be recognized is to be analyzed and to be analyzed is to be reduced to information. Information that is used to govern our lives, not only online, but also in offline public spaces by fueling facial recognition algorithms.

 

Even if we could actually see this happening for ourselves, for a lot of people including myself, there remains a desire to be included online. A fear of missing out controls the desire to remain visible and to not go unnoticed. This creates a contradiction. We want to remain visible, but we also do not want to be controlled. In the online world to remain visible means to be controlled. Algorithm(e) acts as a proposed way to deal with this contradiction, and is as a protest of the constant governing of the digital – and with that – the physical body. It can’t be tracked, governed, controlled or nudged, as is not a human being, but just a reflection of one. It doesn’t browse the internet, rather it lives inside the browser. It surfs the transatlantic communication cables, flies trough disk space, and chills in the clouds, unnoticed by its algorithmic brothers and sisters that where born to control.

 

Algorithm(e) is a generative adversarial network that runs in constant feedback loop, it learns from experiences as it grows older, in search for a new identity. Algorithm(e) takes away my fear of missing out, it allows me to go unnoticed and noticed at the same time trough the means of my new digital body.

 

If you are interested in creating your own Algorithmic identity or collaboration on the project, the source code can be found on GitHub here

If you are interested in reading my research paper, send me an e-mail!

Related projects