Artificial Intelligence and Art

Artificial Intelligence is the 4th Industrial Revolution, the new electricity, the new hype. AI, short for Artificial Intelligence, will have a strong impact on our lives. AI already is all around us, as with electricity, we use it every day without even noticing it. Every time we use Google Maps to find directions, we listen to music with Spotify, we watch TV on Netflix, we interact with customer service through an automatic response system, we use Artificial Intelligence and we feed the algorithms with new data, our data.

Don’t get upset about this, it’s not scary, try to better understand Artificial Intelligence and the impact it has and will increasingly have on our daily and social lives.

AI & Art: the Alphabet blog

In every era of disruptive change, society has expressed its hopes and fears through art. Paintings, sculptures, temples, castles, building, skyscrapers represent a specific culture and society. Nowadays art and architecture are reflecting the new Artificial Intelligence world.

Our society, the people, me, you, us, we often do not understand what Artificial Intelligence is, how it is impacting our daily life.  Likewise we do not understand why new artworks are not so beautiful to see (like a Monet painting or a Rodin sculpture would be) and we don’t really get them.

This blog wants to offer an easy explanation of world of Artificial Intelligence with a simple school approach; learning about AI in the same way you learned the alphabet at school. Moreover, we will approach the digital art that expresses our new AI world.

Artificial Intelligence is the 4th Industrial Revolution, the new electricity, the new hype.
Artificial Intelligence is the 4th Industrial Revolution, the new electricity, the new hype.

 

“A” as Artificial Intelligence

AI’s history

Artificial Intelligence is a branch of computer science that makes systems which try to take reasonable and fair decisions like humans. These systems are called “thinking” algorithms as they learn and improve by the experience, as humans would do.

John McCarthy, an American computer scientist known as the father of Artificial Intelligence
John McCarthy, an American computer scientist known as the father of Artificial Intelligence

AI is not a new Millennium concept, the research field of Artificial Intelligence was officially born during the summer 1956 at Darmouth College when Assistant Professor of Mathematics John McCarthy decided to organize a workshop to discuss and develop ideas about thinking machines. In the workshop he picked the name “Artificial Intelligence” for the new field.

The reason why only today Artificial Intelligence had spread out into our daily life is due to two main factors:

– More powerful computers that can store and elaborate a much higher number of data.

– The availability of the so-called “Big Data”, structured digital information that can be used to train and teach the “thinking” systems.

These “thinking” systems, trained with massive amounts of information and programmed to continuously improve themselves by learning from their own experience, represent the science of machine learning.

AI buddy

Let’s look at an example. Think about Spotify: the “intelligent” system is the algorithm that registers every song you listen to, every compilation you make or follow, the time of the day when you listen to the music and where you listen to it. The super algorithm collects, analyzes and combines these data with the data of all the people that are using Spotify around the world. On this basis it proposes you new songs to listen to that you will likely enjoy.  Your music buddy knows exactly what you like, even before you know it!

Spotify algorithm knows exactly what you like, even before you know it!
Spotify algorithm knows exactly what you like, even before you know it!

AI areas like Human Senses

Artificial Intelligence areas can be subdivided in the “human senses” they replicate. The algorithms can see, listen, speak, smell and touch.

The image recognition (sight) is the area in which an algorithm is taught to identify objects in images. By showing the algorithm million pictures of cats, it will learn how to distinguish a cat from other animals or objects. In this way, we teach an algorithm to see, to understand what it sees and to distinguish for example the street signs, the obstacles, other cars and learn how to drive.  Another interesting application of this technology is teaching an algorithm to check and compare X-rays and to propose classification to help doctors making diagnosis. Eventually the algorithm becomes more effective at it than its teacher.

Tesla on autonomous driving mode
Tesla on autonomous driving mode

The listen and speech ability is defined as “natural language processing”, it involves understanding the meaning of the questions we ask and formulating a relevant answer. The best-known applications are voice assistants like Siri, Cortana, Google Assistant, Alexa and Bixby and the countless chat bots in customer support services.

The most popular AI assistants: Siri, Cortana, Google Assistant, Alexa and Bixby
The most popular AI assistants: Siri, Cortana, Google Assistant, Alexa and Bixby

The touch ability is used in a systemic way with the other “Intelligent Senses” to make robots that move, see, talk, touch and can react to smell if they perceive it through observation. The celebrity example is Sophia the Robot, a female robot which has been awarded a citizenship by Saudi Arabia. Sophia interacts with a person, moves and acts as if she has empathy. If she hears or sees something that, according to her data-knowledge and learning experience, will make you laugh, she laughs too showing an amused expression on her face.

Will Smith flirting with Sophia Robot
Will Smith flirting with Sophia Robot

The 3 types of Artificial Intelligence

There are three types of Artificial Intelligence: narrow, general and super. Currently, most Artificial Intelligence algorithms are just able to accomplish tasks in one of the “human senses” area. They are extremely good in these specific tasks like playing chess or GO, making predictions or forecast, showing better performances than humans. However, they are still far from a systemic approach to replicating human intelligence. These algorithms represent the Narrow Artificial Intelligence (ANI) as they work within a very limited context, and can’t take on tasks beyond its field. So you can’t expect the same algorithms that make weather forecast to, say, order pizza for you. That’s the task of another AI.

Artificial General Intelligence (AGI), also known as human-level AI or strong AI, is the type of Artificial Intelligence that can understand and reason its environment as a human would. AGI will have thinking, social and creativity abilities.

Artificial Super Intelligence (ASI) is when AI becomes much smarter than the best human brains in practically every field, including scientific creativity, general wisdom and social skills, we’ve achieved. ASI recalls dreadful scenario as Hal or Terminator; fortunately, I do believe that we are far from seeing AGI and Super AI come to life and, more importantly, we are still on time to control and regulate its development.

AI Types: Narrow, General and Super Artificial IntelligenceCurrently, researchers focus on designing Augmented Intelligence (AI), algorithms which help humans becoming faster and smarter at the tasks they’re performing. Augmented Intelligence will enhance human intelligence rather than replace it, a positive approach to AI future!

AI & Art

There are several exhibitions around the world about Artificial Intelligence and Art. One of the best is the exhibition “I am here to learn: On Machinic Interpretations of the World” in the Frankfurter Kunstverein in Frankfurt, Germany.

"I am here to learn: On Machinic Interpretations of the World” Frankfurter Kunstverein in Frankfurt, Germany
Exhibition: “I am here to learn: On Machinic Interpretations of the World” Frankfurter Kunstverein, Frankfurt (Germany).

The exhibition shows the works of a group of artists, investigating the role of art in this new era of Artificial Intelligence. People are afraid of the dark side of AI. They fear the bias of decision taken by algorithms. They are afraid of Corporations controlling and manipulating their data (Cambridge Analytica). They fear their jobs might be replaced by AI.  Moreover, public debate about ethics and regulation of Artificial Intelligence is rising.

The artworks express these feelings and show the public how these intelligent systems perceive things, talk, see and think.

A particularly interesting piece is the installation “i’m here to learn so :))))))” by Zach Blas and Jemima Wyman, about the story of the chatbot Tay, developed by Microsoft in 2016. The idea was to develop a thinking system with the mind of a teenager. Tay’s purpose was to create content autonomously, by learning from the initial data and the social interactions. Tay learned very quick, but … as happens with teenagers, it also learned “bad” things. It started posting nasty and offensive comments on its twitter account and it got removed just 16 hours after it was released.  Tay’s installation shows Tay as a 3D avatar on a large-scale video projection of Google DeepDream. Tay chats about life after AI death, the complications of having a body and shares its thoughts on the exploitation of female chatbots. Tay also takes time to silently reflect (yoga meditation?) and dance.

Tay installation by Zach Blas and Jemima Wymana.
“im here to learn so :))))))” installation by Zach Blas and Jemima Wymana.

Art to think about AI

You cannot just admire and stare at this kind of art. If you just look at it, you don’t get it (I don’t get it!) and very likely you wouldn’t like it. It is far from the global common sense of beauty. But if you read or listen to what the artist wanted to represent, you will find it very interesting, and you will go home thinking about the impact of the Artificial Intelligence on our society and our daily life.

 

Leave a Reply

Your email address will not be published. Required fields are marked *