Will AI replace artists
“Will AI replace artists?” Image above generated with Stable Diffusion 2.1, using prompt “contemporary painting.“

We asked two popular AI-powered text tools on the market, Bing Chat by Microsoft and GPT-4 by OpenAI, about the future of fine art artists. Upon fact-checking, we found that most quotes and numbers were not correct; therefore, they have been removed or edited to maintain accuracy according to the original source. The broader answers have not been changed, the only edits made were to the referenced quotes and statistics.*

Will AI replace artists?

Bing Chat: Artificial intelligence (AI) is a technology that can perform tasks that normally require human intelligence, such as recognizing images, understanding language, and generating content. In recent years, AI has been used to create various forms of art, such as paintings, music, poetry, and even movies. Some of these AI-generated artworks have been praised for their creativity and originality, while others have been criticized for their lack of authenticity and ethics.

On the other hand, AI can also be seen as a threat that can undermine the value and meaning of art. AI can potentially copy, manipulate, and exploit the works of human artists without their consent or recognition.

But can AI really replace artists? The answer is not so simple. On one hand, AI can be seen as a powerful tool that can enhance the artistic process and expand the possibilities of expression. AI can help artists explore new styles, techniques, and ideas, as well as collaborate with other artists across different domains and disciplines. AI can also make art more accessible and inclusive, by allowing anyone to create and experience art without any barriers or limitations.

On the other hand, AI can also be seen as a threat that can undermine the value and meaning of art. AI can potentially copy, manipulate, and exploit the works of human artists without their consent or recognition. AI can also create artworks that are indistinguishable from those made by humans, raising questions about the authenticity and originality of art. Moreover, AI can challenge the role and identity of artists, by reducing their agency and creativity in the artistic process.

Therefore, the question of whether AI will replace artists is not a binary one. Rather, it is a complex and nuanced one that depends on how we define and understand art, how we use and interact with AI, and how we balance the benefits and risks of AI for art. Ultimately, AI can be both a friend and a foe for artists, depending on how we choose to embrace it or resist it.

GPT-4: As of my knowledge cutoff in September 2021, AI is not expected to replace artists. Here are a few reasons why:

1. Uniqueness and Human Touch: Each artist has a unique style and perception of the world, often reflecting personal emotions and experiences. This uniqueness and human touch cannot be exactly replicated by AI. AI can create art based on patterns it has learned, but it cannot provide the same depth of human experience and individual creativity.

2. Understanding Context: AI may struggle to understand the cultural and social context that often influences art. Art is a form of expression that often reflects the times, including political, social, and cultural environments. This level of comprehension is currently beyond AI capabilities.

3. Emotion and Empathy: AI lacks human emotions and cannot empathize with human experiences, which are often crucial components of art.

4. Intentionality: Art is often about intention, about a message or feeling the artist wants to convey. As of now, AI doesn’t possess intentions in the way humans do.

However, AI can certainly assist artists and open up new avenues for artistic exploration. For instance, AI can generate interesting patterns, blend styles, and provide tools to help artists realize their visions. Moreover, AI itself can be used to create unique kinds of art, where the algorithm’s role is part of the creative process. This has given rise to a new field called AI Art or Computational Art.

So while AI can’t replace artists, it certainly can be a tool for artists to utilize and a medium for them to explore. As of now, the idea of AI replacing human artists remains in the realm of science fiction. By 2023, there may be developments and advancements, but it’s unlikely that the fundamental elements that make human art unique have changed.

Who buys fine art?

GPT-4: A wide range of individuals and organizations buy fine art, often with diverse motivations. Here are some typical buyers:

1. Private Collectors: Individuals with a passion for art often buy fine art to add to their personal collections. They may be drawn to a specific artist, genre, style, or period, or they may appreciate the potential investment value of the artwork.

2. Investors: Some people buy fine art primarily as an investment, hoping that the piece will appreciate in value over time. These buyers often have a deep understanding of the art market and trends.

3. Corporations: Many corporations acquire fine art for their offices or as part of a corporate art collection. This can enhance the workspace environment, reflect the company’s brand or values, and sometimes serve as an investment.

4. Museums and Galleries: Museums and art galleries buy fine art to add to their collections for public exhibition. They aim to preserve, study, and present artworks for cultural, historical, or educational purposes.

5. Art Dealers and Art Advisors: Art dealers, brokers, and advisors often buy fine art on behalf of their clients. They use their expertise to help clients build collections or invest in art.

6. Government Institutions: Governments might buy fine art for public museums, cultural institutions, or government buildings. This can serve cultural preservation, public education, or diplomatic purposes.

7. Interior Designers and Architects: These professionals often buy fine art to complement the design of a space. Artwork can enhance the aesthetic appeal and create a specific mood or ambiance in a space.

8. Nonprofit Organizations: Nonprofit organizations often buy fine art for fundraising purposes. Art can be auctioned off to raise funds for specific causes.

Remember, fine art isn’t exclusively accessible to the wealthy or educated. Anyone can purchase fine art based on their personal interest, budget, and the value they derive from owning or displaying the artwork. There are artworks available at all price points, from emerging artists to established names in the industry.

What did most fine art sell for in the last couple of years?

Bing Chat: The cost of fine art can vary widely depending on many factors, such as the type, quality, rarity, and popularity of the artwork, the artist, the medium, the size, the condition, the provenance, the market demand, and the sales venue. However, one way to get a general idea of how much fine art sells for is to look at the statistics and trends of the global art market.

According to Statista, the global art market generated nearly 68 billion U.S. dollars in 2022, surpassing the figures reported prior to the coronavirus (COVID-19) pandemic and reaching the second-highest value in the last 15 years. The volume of global art sales also experienced an annual increase in 2022.

The US remained the largest market globally with 45% of sales, followed by the UK with 18% and China with 17%. France was the fourth largest market with 7% (source, page 26).

The most popular segment of fine art at public auctions worldwide in 2022 was post-war and contemporary art, which accounted for 54 percent of the total sales value (source, page 183). The second-most popular segment was modern art, which accounted for 22 percent of the total sales value. The third-most popular segment was impressionist and post-impressionist art, which accounted for over 18 percent of the total sales value.

Another trend that emerged in 2022 was the rise of non-fungible tokens (NFTs) as a new form of digital art. NFTs are unique and verifiable digital assets that can represent artworks, collectibles, or other items.

Going back to who buys fine art, what percentage of all people that buy fine art are private collectors?

Bing Chat: The percentage of fine art buyers who are individual collectors is not easy to determine, as there is no official or comprehensive data on the composition and behavior of the global art market. However, some sources suggest that individual collectors are a significant and influential group of buyers who buy art for passion, investment, or philanthropy.

According to the Art Market 2023 report by Art Basel and UBS, private collectors accounted for 72 percent.

According to a report by Knight Frank, fine art and other collectibles owned by ultra-high net worth individuals (UHNWIs) globally account for just 2 percent of total portfolio allocations. 

Therefore, based on these sources, it can be estimated that individual collectors make up a large and diverse percentage of fine art buyers who have different motivations and preferences for buying art. However, the exact percentage may vary depending on the definition and scope of fine art and individual collectors.

Do you have any data on the percentage of art buyers who purchase art for passion versus other motivations?

GPT-4: Numerous surveys and reports on the art market have suggested that a significant portion of art collectors do buy art primarily for personal enjoyment or passion, rather than purely as an investment.

For example, the UBS and Art Basel Art Market Report from 2017 noted that “enjoyment” and “passion for art and collecting” were the primary motivations for high-net-worth collectors surveyed in the report. The study indicated that 71% of collectors said they bought art for enjoyment, and 54% said they did so because of a passion for art and collecting. (source)

Knowing that private collectors buy fine art mostly for enjoyment and passion, do you think buyers could also enjoy digital art as much as real, tangible art?

GPT-4: Absolutely, many people can and do enjoy digital art as much as traditional, tangible art. Art appreciation is a highly subjective experience and depends on individual tastes, interests, and experiences. 

Digital art has opened up new avenues for creativity and expression, allowing artists to experiment with different techniques, media, and concepts. It can offer a unique aesthetic and interactive potential that isn‘t possible with traditional art forms. For example, digital art can incorporate motion, interactivity, and multimedia elements, and it can be easily shared and viewed across different digital platforms.

However, the tactile nature of traditional art—the physicality of a painting or sculpture, the brush strokes, the texture of the paint or the material, the three-dimensional aspect—can give it a depth and a presence that digital art often can’t replicate. These physical aspects can contribute to the enjoyment and appreciation of traditional art.

Both digital and traditional art have their unique qualities and appeal, and neither is inherently superior to the other. It’s also important to note that the lines between digital and traditional art are increasingly blurred, with many artists combining elements of both in their work.

That said, the market for digital art, including AI-generated art, is still evolving. While some collectors have embraced digital art, others are more hesitant, often due to questions about authenticity, longevity, and the lack of physical presence. Issues like how to properly display, preserve, and authenticate digital art are still being worked out.

In conclusion, whether someone enjoys digital art as much as traditional art will depend on their personal tastes, their openness to new forms of art, and how they like to experience art. As with all forms of art, the key is to find what resonates with you personally.

* Chat AI bots, like any AI system, rely on the data they have been trained on to generate responses. In the case of language models such as GPT-3.5, they have been trained on a wide variety of text from the internet. While efforts are made to ensure accuracy during training, these models may sometimes generate responses that appear to be factual but are actually made up or incorrect. (Source: Chat GPT-3.5, May 2023)