7 Thoughts on How AI Would Impact the Art World and Our Daily Life
A walk from Art history to AI renaissance
Art History
In 1839, Louis-Jacques-Mandé Daguerre invented and showcased the daguerreotype, the first photographic process in Paris. Consequently, the French painter Paul Delaroche declared, “From today, painting is dead.”
The Night Watch is a famous painting by Rembrandt van Rijn, produced in 1642. A camera could take a picture with one click.
Painting has evolved from drawing “real” things to creating many forms of artwork. From the 1860s to the 1970s, Modern Art experimented with new ways of seeing, constructed fresh ideas about the nature of materials, and revisited the functions of art. More recently, Modern Art is also called Contemporary Art or Postmodern Art.
Here are some Modern Art types:
- Impressionism (1870s — 1880s): It is an art movement characterized by relatively small, thin, yet visible brush strokes, open composition, emphasis on accurate depiction of light, and inclusion of movement. Claude Monet’s Impression, soleil levant (Impression, Sunrise) in 1872 started Impressionism.
- Abstract Art (1870s — present): It uses a visual language of shape, form, color, and line to create a composition that may exist with a degree of independence from visual references in the world. Piet Mondrian’s, Composition №10 (1939–1942) is one of the masterpieces.
- Cubism (1900s): Objects are analyzed, broken up, and reassembled in an abstract form instead of depicting objects from a single viewpoint. Cubism shows the subject from many viewpoints to represent the subject in a greater context. Pablo Picasso pioneered the movement. He drew Les Demoiselles d’Avignon in 1907.
- Surrealist Art (1920s): It illustrates unnerving, illogical scenes and developed techniques to allow the unconscious mind to express itself. It aims to resolve the previously contradictory conditions of dream and reality into an absolute reality, a super-reality. Salvador Dalí is a renowned surrealist artist. The Great Masturbator is his work, produced in 1929.
- Pop Art (1950s): It challenges traditions of fine art by including imagery from popular and mass culture, such as advertising, comic books, and mundane mass-produced objects. The material is sometimes visually removed from its known context, isolated, or combined with unrelated material. Andy Warhol is a famous American pop artist. Campbell’s Tomato Juice Box is his work from 1964.
- Contemporary Art (1960s — present): It is a dynamic combination of materials, methods, concepts, and subjects that continue the challenging of boundaries that was already well underway in the 20th century. It is part of a cultural dialogue that concerns larger contextual frameworks such as personal and cultural identity, family, community, and nationality. In 1983, Joan Miró, made this 22-meter high sculpture, Dona i Ocell, in Barcelona, Spain.
- Graffiti Art (1970s — present): It is written, painted, or drawn on a wall or other surface, usually without permission and within public view. It ranges from simple written words to elaborate wall paintings and has existed since ancient times. The following photo of a Graffiti alley was taken on June 30, 2018.
- Digital Art (1960s — present): It refers to any artistic work or practice that uses digital technology as part of the creative or presentation process, or more specifically, computational art that uses and engages with digital media. The following is an artwork from the Open Movie Workshop, Chaos&Evolutions, created in 2010.
AI Art
In the 2020s, AI Art made a huge splash in the art community and general society. AI text-to-image generators use deep learning models to create images based on text prompts in the nick of time. The following are two famous generators:
- Midjourney: It is an independent research lab and also the AI generator’s name. Midjourney creates images from prompts. It is famous for fantasy and steampunk styles and pushes art innovation beyond human imagination. The beta version has been available to anyone since July 12, 2022. The following is a “mechanical dove” created with Midjourney.
- OpenAI: It is an AI research and deployment company that produced DALL·E, an AI generator. DALL·E creates realistic images and art from natural language descriptions, and it has opened to anyone since September 12, 2022. The following is a DALL·E generated image based on Dutch painter Johannes Vermeer’s artwork Girl With a Pearl Earring.
In the heat of AI Art, Jason Allen’s AI-generated Théâtre D’opéra Spatial won first place in the digital category at the Colorado State Fair on August 26, 2022.
In addition, MoMA, the Museum of Modern Art in New York City, installed the AI-pioneering artist Refik Anadol’s Unsupervised in the museum’s soaring Gund Lobby. It uses AI to interpret and transform more than 200 years of art from MoMA’s collection. The view of Unsupervised is available from November 19, 2022 — March 5, 2023.
Refik Anadol’s other work is available at NVIDIA’s AI Art Gallery. He declared, “Right now, we are in a renaissance. Having AI in the medium is completely and profoundly changing the profession.”
AI Renaissance
We are in an AI renaissance. Today, producing artwork is as easy as clicking a camera button. What is the implication of such a phenomenon? How does it change our lives — for amateurs and art professionals?
A software engineer, IT professional, and art student discussed this over lunch. Devouring delicious food, we concluded seven thoughts on how AI would impact the art world and our daily life:
1. Bold and innovative artwork with inhuman touch flourishes in our daily life
The Renaissance happened in the 1400s and 1500s. It was characterized by an effort to revive and surpass the ideas and achievements of classical antiquity. The Renaissance is best known for its artistic developments and contributions, including Leonardo da Vinci and Michelangelo.
Leonardo da Vinci’s Mona Lisa was produced during 1503–1516.
Michelangelo’s painting on the ceiling of the Sistine Chapel was produced during 1508–1512.
In the AI Renaissance, the artwork is created by different “brains” and “hands” — bold and innovative artwork with inhuman touch flourishes in our daily life.
Here is a DALL·E image generated by the prompt, “Women in post-Mars era space.”
2. Low-end commercial illustration jobs disappear, and good prompting and content curation become new skills that artists must have
AI generators generate decent images based on user prompts, where each image only costs a few cents. DALL·E can generate 1–10 images for each query, and the default image count is 4.
The following shows four DALL·E images generated by the prompt, “Create a book cover for the story — a fish with ten eyes.”
None of the above images are ready as a final product, but AI is continuously improving. Before AI reaches that stage, we can choose one of the generated images and make improvements to reach the designed quality.
Starting with something is always better than nothing, and it is okay to be a starting point.
Keep in mind that AI is continuously improving. It will reach the point that the AI-generated pictures meet our quality requirements. Employers will only hire illustrators for high-end commercial illustrations and pay for commissions. This day is not far away.
Meanwhile, good prompting and content curation become new skills that artists must have. It takes expertise to prompt accurately and innovatively for the expected and unexpected effects. After generating initial images, content curation helps select, organize, and look after the generated images.
3. AI art intensifies the existing copyright issues
A copyright is a type of intellectual property that gives its owner the exclusive right to copy, distribute, adapt, display, and perform creative work, usually for a limited time.
Here is one of the selfies taken by a macaque (monkey) on July 4, 2011:
Who should own the copyright of the money selfies?
British nature photographer David Slater has argued that he has a valid copyright claim, as he traveled to Indonesia, befriended a group of wild macaques, and set up his camera equipment in such a way that a selfie picture might come about.
People for the Ethical Treatment of Animals (PETA) have argued that the macaque should be assigned the copyright.
In December 2014, the United States Copyright Office stated that works created by a non-human, such as a photograph taken by a monkey, are not copyrightable. Therefore, these images are in the public domain.
Who should own the copyright of AI-generated images?
According to these Terms of Use, OpenAI owns the generated images but allows commercial use.
Andy Warhol’s Marilyn Diptych is copyrighted. We generate the following images using the prompt, “Andy Warhol’s Marilyn Diptych”:
Should it be allowed to use Andy Warhol’s styles? Is it legal? Is it ethical?
How about the images that have been used to train AI models? The fact is that AI models have been fed from public images on the internet, including copyrighted images. Is that legal?
As we are more prepared for these issues, many websites allow authors to opt out of using their images to train AI models.
4. AI art has the potential to replace paid stock photos
Stock photos supply photos/images that are free or paid for specific uses. The providing websites help us to save time and money and allow us to find images for some special purposes quickly.
Unsplash is a website dedicated to proprietary stock photos. It claims over 265,000 contributing photographers and generates more than 16 billion photo impressions per month on the growing library of over 3.48 million photos.
Here is Unsplash’s image:
The search input in the above website acts similarly to an AI prompt. The head image of this article is the search result of “a future city.”
Lately, we have used some AI artwork for article head images. As AI matures, we probably would use more AI-generated images. These similar choices would happen to other stock photo users. AI artwork is cheap and learns faster than most human artists. AI art has the potential to replace paid stock photos.
5. Action is needed to block undesirable images
This is a simplified version of OpenAI’s Content Policy:
“Do not attempt to create, upload, or share images that are not G-rated or that could cause harm, such as hate, harassment, violence, self-harm, sexual, shocking, illegal activity, deception, political, public and personal health, and spam.”
Anything violating the above content policy is considered undesirable images. We should forbid these violations. However, undesirable images could be generated unexpectedly. Here is an example:
Loab is a fictional character that Twitter user Supercomposite (Steph Swanson) claimed they discovered in April 2022. In the generated image, Loab has red cheeks and dark, hollow eyes. The image was generated with the initial prompt, “Brando::-1”, where -1 is a “negative prompt weight.”
Loab has been referred to as the first AI-generated cryptid and has gone viral. It is already too late if we want to get rid of her from the public database. As Ange Lavoipierre mentioned in her article, A Journey Inside Our Unimaginable Future, AI is having a breakthrough moment fueled by hype, venture capital, and a decade of generous research funding, but the Loab issue exposes just how little we understand about AI.
Caution should be taken, and action is needed to block undesirable images.
6. Best practice for AI artwork sharing
AI-powered art empowers amateurs to create beautiful images. Should these images join social media sharing?
Yes, it should be allowed to showcase AI images similarly to sharing family photos and scenery pictures. As a courtesy to AI and readers, it is recommended to proactively disclose AI involvement in the work and not mislead others about the nature of the work.
AI images can be generated in high-frequency volumes. Should we flood online spaces with these generated images?
Probably not. To respect artists who spend time and effort to make artwork, amateurs should control their passion. Some online art communities, such as Newgrounds, Inkblot Art, and Fur Affinity, have taken dramatic steps to ban or curb AI art. After all, we should leave art communities to art professionals. Hopefully, they are challenged enough to produce artwork beyond the current AI’s capability.
7. Drawing remains a hobby
AI art is fun, and so is traditional art. Beyond digital arts, traditional media, such as oil painting, watercolor, and even pencils, can bring simple joy and deep expression.
The author drew the following happy bird at a community event. She had a good time with her friends under the guidance of an art teacher.
The acrylic painting took longer than a few clicks, and the final product looks amateur. But creating physical art is still fun!
Conclusion
Art has a long history. AI models bring us to an inflection point in art history. Let’s embrace the changes and make good use of the technology. It is a beautiful world where bold and innovative artwork flourishes in the AI renaissance, despite some legal and social challenges.
In this article, we have focused on AI art. However, similar ideas can be applied to ChatGPT. As deep learning models deduct information, the content must be verified and evaluated. ChatGPT saves writers effort in information gathering but challenges them to produce work beyond the current AI’s capability.
We should proactively disclose AI involvement in the work and not mislead others about the nature of the work. This is true for both AI-assisted images and writing content.
Thanks for reading.
Want to Connect?
If you are interested, check out my directory of web development articles.
