Art and artificial intelligence: simply a tool or an ingenious creation?

A study shows how language humanizes artificial intelligence

Paintings, poems, and music are all created using intelligent algorithms. International researchers (from MIT and the Center of Humans and Machines) have found that people perceive artificial intelligence (AI) as an ingenious creator of art.

Similarly, art and Artificial intelligence may simply be another tool used by artists, depending on how the information is presented. iScience published the study’s findings.

Using an intelligent algorithm, an artwork by Edmond de Belamie was created and auctioned for 432,500 USD at Christie’s Auction House in October 2018. It was created by artificial intelligence (AI), according to Christie’s auction advertisement.

A machine was described as the first to create art without being guided by a human.

As a result, the machine didn’t receive any proceeds, but instead the French artists’ collective Obvious. The collective had incorporated the photos of real paintings by human painters into an algorithm and trained it to create images independently. A picture was selected, printed, given a name, and marketed.

It was unknown whether the programmers who devised the artificial neural networks and algorithms used in the painting received any of the proceeds.

There are many participants in the process of the mixture of art and artificial intelligence, among them artists, curators, and programmers. Additionally, there is a tendency — especially in the media — to endow artificial intelligence with human-like characteristics.

It appears from the reports you read that creative artificial intelligence can automatically produce brilliant artwork. The first author of the study Ziv Epstein, a Ph.D. student at the MIT Media Lab, explains that his study aimed to determine if there is an association between the humanization of Artificial intelligence and the credit given to their work.

About 600 participants were informed about how the mixture of art and artificial intelligence is created and asked who should be recognized for their contribution. The participants also assessed the degree to which AIs are humanized by each other.

Each answer differed greatly. Overall, those who humanized artificial intelligence and did not perceive it only as a tool also thought it should be recognized for its art and not for the people involved in its development.

It was initially the artists who provided the algorithms with data and trained them who were given the most recognition in the process of creating art and artificial intelligence. After that, curators would be appointed, followed by technicians who would program the algorithms.

As a final note, the “crowd” (i.e. the mass of Internet users that produce the data material that AIs are usually trained on) was also mentioned. People who humanized artificial intelligence gave more credit to technicians and the crowd, but proportionally less to artists.

Likewise, when respondents are requested to identify who is responsible for a copyright violation, for example, similar pictures emerge. AIs are also viewed as more responsible here by those who humanized them.

It was discovered in the study that the language used to report on artificial intelligence systems in art can actively manipulate whether people humanize AIs.

Artificial intelligence generates works of art with the help of an artistic collaborator, which is the definition of the creative process.

Another way of explaining the process would be to explain that an artist conceptualizes the artwork, and the AI executes simple commands provided by the artist. Different descriptions led to different levels of humanization and, therefore, different perceptions of who was acknowledging and assuming responsibility for art and artificial intelligence among the human actors.

We will have to pay more attention to who is responsible for what AI creates because AI is infiltrating our society in an increasing manner. Every AI has a human behind it. The problem is especially acute when artificial intelligence malfunctions and causes damage, such as when an autonomous vehicle is involved in an accident.

According to the study’s co-author, Iyad Rahwan, director of the Center for Humans and Machine at the Max Planck Institute for Human Development and a co-author of the study, language affects our view of AI. “Humanization of artificial intelligence contributes to difficulty in assigning responsibility”.

Does tech really count as creativity?

As a result, AI can appear creative if you view creativity as just another skill that can be acquired through data.

Since Artificial Intelligence has been advancing at such a tremendous rate, it’s only natural to expect that in the near future, AI-generated art will be indistinguishable from human art in any field.

However, is this an adequate answer? Objectivity and understanding are fundamental to any concept of “artistic” creativity, and the concept of “artistic” creativity has deep philosophical contentious roots.

Currently, computers cannot produce truly random data due to today’s technology. Moreover, some might argue that randomness is what gives human creativity its spark — adding something no one else has ever thought of or been able to add before. This cannot be achieved by artificial intelligence.

Additionally, AI may be able to convince humans that a poem, painting, etc., was created by another human, but that does not mean the piece of “creative art” actually represents something meaningful. Unless you give an AI input to understand, it will produce an endless number of outputs.

A string of words it analyzed appeared in another piece of text is analyzed, so it compares you to a summer’s day. But it has no idea how that string of words relates to you. The inputs and outputs of artificial intelligence are merely data.

The good news is that humans are innately creative, so that data can be used creatively.

Artificial Intelligence and creativity: Can AI enhance creativity?

Human creativity can be enhanced and strengthened with the help of artificial intelligence.

AI, for instance, is excellent at doing lots of things in a short amount of time. An individual can use it to accelerate and inspire their creative process by coming up with many ideas, concepts, or rough drafts.

Even though AI may not be able to produce the final product, it can act as a co-creator.

A human was involved in any of the examples previously outlined, even if it was just selecting the best artwork created by an AI.

A computer cannot determine what humans like, so it’s still up to creative people to determine what resonates with other people. The best creations can still be improved upon if humans discard the less inspired ones and use their own intuitive abilities.