How New Tech Can Be a Creative Asset
Artificial intelligence (AI) and machine learning (ML) are all over the Internet right now: they are being discussed on TV, radio, Twitter and more. Mostly we hear scare stories about the displacement of jobs, but they actually just represent a new way of working, just as the introduction of machines did during the industrial revolution. In fact they are very powerful tools that arts and crafts people should be making more use of – and in fact much art has already been made using them.
It is worth clarifying though that they are not the same thing and shouldn’t be used interchangeably as terms. I am sure you know this, but I still hear many people using the terms ‘web’ and ‘Internet’ interchangeably and of course, if they still do that then I suspect a similar thing happens with AI and ML.
So what are they and how might they inform your practice?
OxfordDictionaries.com defines artificial intelligence (AI) as being ‘the theory and development of computer systems able to perform tasks normally requiring human intelligence, such as visual perception, speech recognition, decision-making, and translation between languages’.
They define machine learning (ML) as being ‘the capacity of a computer to learn from experience, i.e. to modify its processing on the basis of newly acquired information’.
This is a crucial distinction. The former is about machines being able to perform human tasks and the latter is the process by which it might learn, similarly, the human body has a physical autonomic system and the brain has the capacity to learn.
Of course it is also worth remembering that artificial intelligence is no match for natural stupidity, of which there is an abundance.
Peter Norvig, an AI scientist and a director of research at Google, has come out and said that Google is a machine learning company and that AI will follow from that.
‘Internally we use machine learning more…it’s what we’re going to use to become an AI-first company’.
So, they are using one to get to the other. Their public successes with Deep Dream is well known and the images have become the fractal patterns du jour.
I personally think these images lack deep impact, nothing there has really moved me. I have utter admiration for the technology and techniques but the images feel rather mechanical and processed. However, it is early days and indeed Google themselves recognised this by launching Magenta as an initiative to try to bridge this gap. Go to https://magenta.tensorflow.org/ and you will see that they recognise very clearly this need to get people involved.
This juncture where artists and computers mix is the most interesting point. It is what drew me into making computer art 25 years ago. In reality, ML and AI are just tools, advanced though they are. It’s the addition of the human factor that creates compelling art rather than just interesting images. There is a tendency to expect machines to be the thing that creates but in reality the most interesting results will come from the interface between people and tools, just as it always has: from the invention of oil painting to the development of photography. The ‘wow!’ moment only occurs when people control the technology and produce human artworks that are novel. There is little ‘wow!’ in using computers to reproduce existing techniques.
Out in the Field
AI and ML have had impact outside the standard creative arts field. In advertising, McCann launched an AI creative director to help develop successful ads, the success of which will only be truly seen over time.
There are lots of artists and performers now working in this field. Check out the list here for a starter and primer. You will notice that the works are often cross-genre in the traditional sense, but that they do usually focus on one area of experience: be that performance, audio, visuals or interactivity.
But How Can all This Help You as a Creative?
Well, the notion that ‘the robots are coming’ is overstated. These techniques are just tools that can help you explore new avenues of creativity, in particular ML. You should start to use them and try them out and the effects on your practice may be profound. The tools can be used in any number of ways and need not be the thing that produces work; they might be used to inform work as well as produce it. As an illustrator you might define the styles and let the computer create varied work based on your rules. As a choreographer there is much that can be done with these systems to further integrate movement, light and sound. You become the conductor not the pianist.
These techniques can be used to provide a critical friend or a creative partner. Imagine for instance a traditional creative pair in an agency…except the copywriter is an ML tool that is trained by the art director. Imagine also that you use ML to interact with your audience. I have an unfinished play, an updated take on Frankenstein, that does just that. If I ever find sponsorship I will be building a system that learns and modifies the play.
Of course, one thing I haven’t mentioned yet regards the aesthetic categorisation of such work. All art has, so far, been classified into a taxonomy of aesthetics, but this area is more often than not simply referred to as ‘computer’ or ‘digital’ art, both of which simply describe the toolset. Given the inherently new nature, that collision of genres, the addition of time, interactivity, computation and networking that ML and AI creates, it does seem time to develop the language to make it capable of describing this. And that we have not done.
Maybe Google are working on it, or maybe this could be handed over to ML and AI to solve.
Richard has been creating new digital products, games and art since 1990, but most recently has been designing and leading digital transformation programmes, working with data, AI and enterprise architecture. He has worked in senior roles at The Royal Shakespeare Company, Microsoft Studios, Aviva, BSkyB, BBC and others. He has also founded a university digital arts department, helped launch the School of Communication Arts, held a Visiting Professorship and is currently Visiting Senior Fellow at Lincoln University. A published author, he also writes and releases contemporary music, writes science fiction tales and photographs his environment.