This week, I finally got a peek at the Spring syllabus for an undergraduate course I’m co-teaching. Sadly my students won’t be watching Blade Runner or reading Do Androids Dream of Electric Sheep? this year. I will be teaching a session on ‘the death of the book’, though, and science fiction plays an increasingly important part in this discussion.
Several years ago, Google released strange, surreal pictures its neural network ‘Deep Dream’ had painted from random noise. In an article entitled ‘Yes, androids do dream of electric sheep’, The Guardian described the process as follows:
What do machines dream of? New images released by Google give us one potential answer: hypnotic landscapes of buildings, fountains and bridges merging into one.
They were created by feeding a picture into the network, asking it to recognise a feature of it, and modify the picture to emphasise the feature it recognises. That modified picture is then fed back into the network, which is again tasked to recognise features and emphasise them, and so on. Eventually, the feedback loop modifies the picture beyond all recognition.
Since then, Google has also launched Magenta, which aims to use ‘machine learning to create compelling art and music’. One of its first products was this computer-generated piano variation on ‘Twinkle Twinkle Little Star’ (drum added later by a human):
Early last year, MIT Technology Review‘s Martin Gayford looked at several of these examples of robotically generated art to try and get at the question of what makes art ‘art’ in the first place:
The unresolved questions about machine art are, first, what its potential is and, second, whether—irrespective of the quality of the work produced—it can truly be described as “creative” or “imaginative.” These are problems, profound and fascinating, that take us deep into the mysteries of human art-making.
Computers have broken into the art world, then, but what about writing? There, too, AI has been making great progress. The Verge‘s Josh Dzieza delved into the strange world of computer-generated novels back in 2014, shortly after Google released its ‘Deep Dream’ images:
Narrative is one of the great challenges of artificial intelligence. Companies and researchers are working to create programs that can generate intelligible narratives, but most of them are restricted to short snippets of text. The company Narrative Science, for example, makes programs that take data from sporting events or financial reports, highlight the most significant information, and arrange it using templates pre-written by humans. It’s not the loveliest prose, but it’s fairly accurate and very fast.
To top it all off, you have the trend of super-realist art, or human-made art that itself looks very similar to what these machines are producing. Writing about Juan Geuer’s Water in Suspense, scientist Michael Nielsen describes how this kind of art works:
Water in Suspense reveals a hidden world. We discover a rich structure immanent in the water droplet, a structure not ordinarily accessible to our senses. In this way it’s similar to the Hubble Extreme Deep Field, which also reveals a hidden world. Both are examples of what I call Super-realist art, art which doesn’t just capture what we can see directly with our eyes or hear with our ears, but which uses new sensors and methods of visualization to reveal a world that we cannot directly perceive. It’s art being used to reveal science.
Although I’m not an artist or an art critic, I find Super-realist art fascinating. Works like the Hubble Extreme Deep Field and Water in Suspense give us concrete, vivid representations of deep space and the interior structure of a water droplet. For most of us, these are usually little more than dry abstractions, remote from our understanding. By creating vivid representations, Super-realist art provides us with a new way of thinking about such phenomena.
Regardless of whether we think machines will kill art, or take it to the next level, I’m very much looking forward to bringing these kinds of questions to my first-years.
