ART OF US: The Collaboration of Human Vision and Artificial Intelligence

How do we define artificial? How do we define intelligence? How do we define art?
Even if you had quite comprehensive answers to all of these questions, your answers are unlikely to stay the same in a couple of months in this fast-paced era with path-breaking technological developments taking place one after another. Even though the science behind them is not so easy to understand due to complex procedures and structured multidisciplinary aspects, educating ourselves might be the only solution to waking up perplexed, unaware of how the human race got to such a point when the future becomes ‘today’. As a word, artificial refers to “anything that is made or produced by human beings rather than occurring naturally” and artificial intelligence refers to “the simulation of human intelligence processes by machines.” When we consider art as “the expression or application of human creative skill and imagination”…How does artificial intelligence create art and since when can it do it?

AI’s path to gaining intelligence goes back to the 50s and it is full of important milestones, from the introduction of the terms artificial intelligence and machine learning, development of the Turing Test, to the creation of art through generative systems. Many pioneers made significant contributions to the world of AI and with the development of the first program to autonomously create original paintings called AARON, improvements followed one other rapidly.
AARON was developed in the early 70s by Harold Cohen: a painter who was later interested in computer technology due to his questioning of the mechanics of art on a deep level. His thoughts about the algorithm of creating art, how the process of creating art develops for an artist, what makes art represent specific emotions, and so on, lead him to design AARON.
Initially, AARON could paint through a rule-based generation of images using mathematical patterns and algorithms. It did not have the software that was complex enough to allow it to paint with color, so the images were black and white. Cohen worked on AARON for years, developing it one step at a time. In the final stage, AARON could paint with colors and its works were exhibited in many museums.

In the following years, IDE (Integrated Development Environment) was designed by Casey Reas and Ben Fry. A program that teaches the basics of computer programming to users who have no background in computer science thus, provides information and the opportunity to explore deeply about generative art for a broader community. Images of objects from an immensely wide network were made free for users with project ImageNet by Dr. Fei Fei Lei, aiming to contribute to the improvement of computer vision and deep learning. In 2014, Ian Goodfellow and his colleagues designed GANs (Generative Adversarial Networks). A kind of learning process with the help of which, the machine can generate original and corresponding creations. So the AI could draw the faces of humans who never existed, paint pictures that were never painted before, and create in many other fields. Thus, the design of GANs aided the process of Generative Art to develop greatly, leading to the rapid escalation of advancements in technology.

In 2015, a computer vision program named Deep Dream was created by Alexander Mordvintsev. A program that can detect patterns in images and create hallucination-like visuals, resembling a dreamy effect. The scientific and artistic communities were highly intrigued by the creations of the program. In the same year, NST (Neural Style Transfer) was launched. A class of software algorithms that can blend two images in means of style, approach or color and create something new involving both. In 2021, open AI released Dall-E, a program that can create images from text captions like “the blue footprints of a flying tree“ or “a cave painting in the style of David Hockney”, paving an abstract way for us to think outside the box and rediscover imagination.

In light of all these developments, it’s clear that we are experiencing a technological breakthrough. Humans have developed artificial intelligence through analyzing, processing, and transferring intelligence into machines, blowing knowledge into creation and AI became a major pool of resources and endless data, blowing us the knowledge back to further improve creation. Two different worlds continue to feed each other every single moment in the world we designed, building the blocks of a new reality… Two different worlds that are combined under a mutual phenomenon: intelligence.

If you ask me, no matter where this story goes, everything about it will always represent the laws of nature. The curiosity, the urge, the intelligence, the competence, the generosity, the art…
We cannot say that the next generation Mozart will not be an AI trying to fight for community rights through the expression and freedom of art. Even though all these future concepts sound quite exciting, there is still a few years ahead for the necessary advancements to take place in order for us to meet these cyber projections that are being planted. Buckle up, observe, get involved, or keep on thinking that there isn’t a slight chance of any of these happening. I just know one thing for sure: While the world is accelarating on this ramp, you don’t want to feel a stranger to it.

REFERANCES

https://www.libreai.com/a-short-overview-on-ai-art/amp/

https://en.m.wikipedia.org/wiki/Generative_art

https://en.m.wikipedia.org/wiki/Harold_Cohen_(artist)

https://www.chiswickauctions.co.uk/news-item/harold-cohen–the-first-digital-artist/

https://computerhistory.org/blog/harold-cohen-and-aaron-a-40-year-collaboration/

https://www.oxfordlearnersdictionaries.com