Women of the AI art revolution

With advances in technology having transformative effects on industries worldwide, it was only a matter of time before artists found a way to use technology to innovate and push boundaries within their craft


By Noma Moyo


Earlier this month, Sotheby's made its first foray into artificial intelligence (AI) art with the auction of Memories of Passerby 1, an abstract portrait that is made by an AI ‘brain’ that beams a stream of distorted faces and blends them together. Despite the hype surrounding AI art, the portrait, which was created by computer scientist Mario Klingemann, sold for only £40,000. To many, this is small change in comparison to the AI painting titled Portrait of Edmond de Belamy, which sold for £337,000 at Christie's in October last year. The disappointing price for Memories of Passerby 1 made many breathe out a sigh of relief - the robot apocalypse is not here yet... Or so they thought.


While media coverage surrounding AI art has mainly focused on what is going on in auction houses and debates over who should get credit for art created by computers, the AI art market is slowly growing and is estimated to be worth more than $100m (£77.7m) according to data by Research and Markets.



But what is even more surprising is that although the AI art trend has only gained momentum recently, computer and algorithmic arts pioneer, Vera Molnár, has been exploring how computers can be used to produce art for decades. She created the first code-based drawing programs and iterated combinatorial images from as early as 1959. Granted, she wasn’t using the highly-advanced AI programmes that many artists are using today, she still deserves a mention for exploring a concept that was way ahead of her time.



Today, there is a new wave of women artists incorporating AI into their artistic process, such as Anna Ridler, who uses generative adversarial networks (GANs). She generates her own bespoke data sets that train her AI models to create images from scratch as she believes this process gives her more control over the end result. She has exhibited her work at Ars Electronica, Tate Modern, and the V&A.


Speaking to Artnet about her work, she said:


“The most interesting part of working with machine learning is the way that it repeats your idea back to you, but in a way that is looser and wilder and freer than I could ever make by myself.”


Another artist who is using GANs is Helena Sarin. Sarin studied computer science at Moscow Civil Engineering University and has toyed with the idea of combining her interests in programming and art in the past. But it wasn’t until she had to use CycleGAN, a GAN variant, to generate synthetic data sets for a client that she had the idea to use the programme for her own photography and artwork. She founded Neural Bricolage Studio to demystify, promote and display AI assisted artwork.



Meanwhile, artist Sougwen Chung uses her work to explore the difference between handmade and machine-made marks in order to understand the relationship between humans and computers. For her project Drawing Operations, she used Google’s TensorFlow, an open-source software library used for machine learning, to classify archives of her own drawings. The software then transferred what it learned about Chung’s style of drawing into a robotic arm that would draw alongside her.