As start of the new event series “Hybrid Futures. The Arts and Sciences – past, present and future“, a collaboration between the Hybrid Plattform and the Futurium, we looked at Machine Learning. The event staged Hito Steyerl, Mike Tyka and Jules LaPlace to discuss the immense influence that Artificial Intelligence (or Machine Learning as they prefer to call it) already has on technology, art and society and the ways it might further influence those areas.
The evening kicked off with a short welcoming by the director of the Futurium Dr. Stefan Brandt and Nina Horstmann of the Hybrid Plattform, after which the audience was drawn to Hito Steyerls presentation of her own work. Hito Steyerl, previously honoured as one of the 100 most influential people of the art world, is largely known for artistically and socio-theoretically reflecting upon our world in times of hyper-capitalism, globalization and increasing political crises. This evening, she presented different approaches she takes to question the means by which Artificial Intelligence predicts the future. One of her projects addressing this is her "Political Pyromancer", an artificially generated documentary video stream of a burning campfire looking 0,04 seconds into the future. Steyerl gave a condensed summary of the history of fire, how it was the first symbol for technology; stolen, used and abused by humans from the immortals up on Mount Olympus. How fire shaped humans into what we are today, creating society by enabling us to cook and keeping us warm while at the same time also shaping the environment surrounding us. What cannot be forgot is that divination with fire has a long history; pyromancy is looking at the future through flames.
There is no knowledge on what AI is going to look like in the future, but fire opened up the night for human activity. It allowed for a colonization of time, similar to how big data based predictions (knowledge gained through Machine Learning) have become a new type of time colonization. Predictions of the future are based on past training data, meaning the past is colonizing the future. This leads to the conclusion that whoever owns this data from the past subsequently colonizes the future.
The parameters of her "Political Pyromancer" are crucial in determining the outcome but once those are set anyone can predict their own future, even if the predicted future is already in the past when you're able to see it.
Next up on the panel was Mike Tyka who presented his kinetic installation "Us and Them". He trained a model on a recently released set of two hundred tweets from accounts identified as bots that were banned after the 2016 US presidential election. Artificially generated Twitter profiles connected to 20 live printers endlessly spewed out machine-generated political tweets. A torrent of political propaganda in the shape of descending curtains of paper surrounds two chairs underneath the elevated structure of printers. It is an investigation into doublespeak and fake news; synthethic media and the blurring of their boundaries of what is real.
Last but not least did Jules LaPlace showcase an image-, language- and sound-generating tool meant to illuminate the creative potential of future technology. He also presented "VFRAME" that started as an open-source prototype utilizing computer vision's ability to detect objects and locate illegal munitions within vast amounts of video from conflict zones provided by the Syrian Archive. For the project, a visual search engine and custom algorithms for processing million-scale video datasets were developed. The partaking members of the project hope to enhance the discovery and documentation of war crimes and human rights violations. The project was granted an Award of Distinction in Artificial Intelligence at Ars Electronica in 2019.
After the three speakers finished giving account on their projects relating to the topic at hand, they discussed among themselves and with the interested audience questions around the way society is being trained to react to machines and their sometimes emerging bias.
We hereby thank all participants and hope to repeat such an ingenious evening again – sometime in the future.