Can Artificial Intelligence Replace Creative Artists? - An Opinion

 By Aliol

Physics Department

We have witnessed with the rise of IoT that artificial intelligence is replacing many human jobs. Drivers are been replaced by self-Driving cars, Pilots are been replaced by auto-pilots and so on. It is just a matter of time when AI will completely replace these jobs. In the long run, no job will remain safe from the rise of AI. Even Creative artists such as poets, composers, musicians, singers etc. should be put on notice. Many people claim that AI is good at various things, it excels way beyond human expectations, but it will be challenging to exhibit creativity as it requires human intuition which AI completely fail to mimic. Or is it possible to mimic human intuition? In the last few Decades, research in areas such as neuroscience and behavioural science allowed scientists to hack humans and, to gain a much better understanding of how humans make decisions. It turned out that our choices of everything, from food to mates result not from some mysterious freewill, but rather from billions of neurons calculating probabilities within a split second. They concluded ‘Human Intuition’ is in reality 'pattern recognition'. Good drivers don’t have exceptional intuition about traffic; rather by adapting and recognizing recurring patterns they avoid careless pedestrians. Moreover, this biochemical algorithm of the human brain is not perfect and make stupid mistakes sometimes resulting in many road accidents in the world. This simply means that AI can outperform humans even in jobs that require human intuition. Here AI is not competing with the mysterious human souls with consciousness about emotions but instead the biochemical algorithm of the human brain which is far from perfect as stated already.

 In today’s world, art is usually associated with human emotions and we tend to judge art by its emotional impact on us humans. Moreover, emotions are not some mystical phenomena but are the result of biochemical processes. Hence, soon, a machine learning algorithm could analyze the biometric data streaming from sensors on and inside your body, determine your personality type and your changing moods and calculate the emotional impact that art is likely to have on you. Of all the different forms of art, music is the most vulnerable to bigdata analysis because both inputs and outputs are just mathematical patterns. The inputs are the mathematical patterns of sound waves and the outputs are the electrochemical patterns of neural networks.

Suppose if you had a fight with your girlfriend. The AI in charge of your sound system will immediately get to know about your inner emotional turmoil from smart wearables that you have put on. And based on what it knows about your personality and human emotions in general will play you, songs that resonate with your distress. After which it can play a cheerful song that might not work on others but will work on you as your subconsciousness connects to a happy childhood memory that even you are not aware of.  No human DJ could ever hope to match that skill of AI.

Eventually, if your girlfriend dumps you, the AI may walk you through your grief, first helping you deny what happened by playing Bobby McFerrin’s ‘Don’t Worry, Be Happy’, then whipping your anger with Drake’s ‘Take Care’, encouraging you to bargain with Taylor Swift’s ‘All Too Well’, dropping you into the pit of depression with Adele’s ‘Someone Like You' and ‘Hello’ and finally helping you to accept this situation with Justin Timberlake’s ‘Can’t Stop the Feeling’. 

The second step for the AI would be to tinker with the songs and melodies, changing them slightly because you dislike that portion in the song. The AI knows because your hormone level drops whenever that annoying part comes. In the long run, AI can learn to compose the entire music with your personalized touch. However personalized music may not be hit as people will continue to prefer the common music that everybody likes, to dance and sing together. But with huge biometric data of the whole population, it would be very easy for AI to compose the music and melodies that everybody likes. It will not be surprising if AI wins Grammy recurrently.

References:

1.    1. Daniel Kahneman, Thinking, Fast and Slow (New York: Farrar, Straus & Giroux, 2011)

2.   2. Wu Youryou, Michal Kosinki and David Stillwell (‘computer-based personality judgment are more accurate than those made by humans’, PANS, vol 112, 2014 )