After years of development and an existence mostly limited to niche, technological audiences, AI is having its first bona fide mainstream moment. Following the attention-grabbing launch of OpenAI’s ChatGPT and the recent virality of Lensa’s AI selfie app, the excitement surrounding AI has crossed an important threshold in our social consciousness. The vast potential of their impact is beginning to be understood.

Naturally, these developments have engendered, alongside excitement and anticipation, disdain and scepticism. Typical of responses to new technologies, negative reactions in the cultural sphere stem from concerns over the risk to existing artistic careers. Specifically, there are major objections to the idea that artwork could be incorporated into AI models without compensating the original creator, whilst creating an instant alternative to highly intentional pieces of art. It is an interesting collision of the tech world’s at times open-source culture, and artists who know the pain of being exploited all too well.

Those concerns are valid: on GitHub, code might be open-source, but attribution is always clear, whilst much of the tech world is decidedly closed. Fortunately, dystopian projections of an AI future are not the only ones we have to ahead of us. Indeed, many artists speak to the opportunities that artificial intelligence offers their craft. But what is indisputable is that as AI creative tools proliferate, the ways in which we view and value the art they help create will fundamentally change.

In attempting to anticipate how, history provides interesting parallels. Certainly, when it comes to AI, the arguments against it mimic a common refrain. A new kind of technology emerges that provides new artistic tools. Artists working in existing mediums feel threatened by the new one, which they label the antithesis of art, rather than its evolution. Just about every genre of visual art or music in the modern era has faced similar accusations. 

“There is a fundamental misunderstanding about what creative AI is and can be.”

— Sasha Stiles 

For one of the highest profile examples, one need look no further than the esteemed poet and critic, Charles Baudelaire. Though it seems preposterous in today's world, Baudelaire and several of his contemporaries took spectacular umbrage to the concept of photography during the mid-19th century. Baudelaire labelled the camera as “the refuge of every painter too ill-endowed or too lazy to complete his studies,” whilst he expected that in his home country, photography would “contribute much to the impoverishment of the French artistic genius.”

Whilst Baudelaire anticipated that photography would bring much change, he did not anticipate what it would offer, scarcely imagining a society, as ours is today, that is overwhelmingly indebted to the camera. We now know that photography has far surpassed the expectations of its critics. It is not unreasonable to suggest that the same will happen with AI art tools. Indeed, it already is. As good as DALL·E 2 and Midjourney are, it is rare that people enter an ill-thought prompt and get a masterpiece in return.

Indeed, poet and actor Laurence Fuller notes that he spent months making more than 4,000 outputs through the AI image generator Midjourney before he ever used a single one in his work. “It takes a lot of determination to work through threads of outputs to really refine what you want to get out of the piece,” Fuller observes. And whilst he notes that AI does make it much easier to get to an outcome that is “good enough”, the many AI artists who are serious about their craft are regularly going far beyond the bare minimum effort that is required.

This is also true in the world of AI text generation, as the poet Sasha Stiles points out. Stiles doesn’t see her AI-assisted work as easier than her other work; rather, it is a different process with different challenges and intentions.

In some ways, it is ironic that artists who use these tools should be criticised for incorporating AI art into their process. After all, it is the traditional art world that first set the precedent. It has often been the case that one specific artist has been celebrated for the work that countless unnamed humans have done. Many of the Renaissance masters had studios in which they simply directed the work of other artisans, as opposed to putting brush to canvas, or chisel to marble, with their own hands. Warhol's Factory was no different.

As Fuller notes, some of our own modern auteurs act as precursors to this sort of collaborative work. In filmmaking, the director is the leading visionary who takes the brunt of the credit, or blame, for a film. And yet a director does not operate the camera, rig the lights, create the sets, or design the visual effects. They also tend not to appear in front of the camera, and they frequently shoot scripts that they did not write. Are film directors any less the artist because they are not physically producing every part of their artwork?

With this in mind, the art of directing provides a helpful case study to consider how our views on art will change alongside technology. Film is probably the only mainstream example of an art form where the most valuable skill is not a tactile one, but an abstract one. We can watch an artist’s drawing style improve over time, sketch by sketch, but we cannot observe a director’s sense of taste, vision, and storytelling in the same way. We can only view the end product, and say whether it made an impact on us or not.

An even closer, though much less popular, comparison is contemporary art. Artists and tastemakers in that realm have long embraced concepts of art that the average person can, and frequently does, find laughable — a prime example being the banana on the wall at Art Basel 2019. Clara Peh, founder of NFT Asia and an established art curator, explains how her perspective made it easy to embrace the work of the AI artists that she curates.

“In the contemporary art discourse in general, there's been a lot more emphasis on intentionality, over effort or skill,” she explains. “There are a lot of interesting works in the contemporary art canon that didn't necessarily require skill in the same ways that Renaissance paintings did. But they were able to achieve certain things. They were able to spark certain conversations. They were able to meet the artist’s intention in their own ways. And I think those works are also very effective in their own right.”

The biggest obstacle in learning to value new AI works is the underlying conceptual squeamishness that we have towards artificial intelligence. People love the idea that art is an intrinsically human activity, and that any art that bears the contribution of a machine is lacking in that biologically-derived soul. Any artist actually using the technology, however, will pitch a very different vision: one which sees AI as an extension of human creativity, not a replacement for it.

“There is a fundamental misunderstanding about what creative AI is and can be,” argues Stiles. “It's not like I'm replacing myself as a writer or outsourcing myself. I'm shifting my attention to different parts of the writing process and utilising tools to craft language in different ways.”

Fuller adds that AI allows him to do completely new things, rather than being tied to that which is familiar. “The ability that it has given me, ultimately, in the work that I do, is a sort of limitless potential to tell whatever story I want to tell with those two minutes that I have to tell the story of a poem. I can go anywhere now. And that has been really exciting.”

AI is only going to become a bigger part of the cultural conversation over time. It has already created the potential for artists to explore and innovate in entirely new ways; by welcoming these technologies, artists could be early pioneers in this brave, new world, one that will need no small amount of explanation.

“I'm most interested in works that are somewhat self-reflective of the technology that they're using. The work can help form a dialogue that asks these wider questions,” Peh explains. For the artists who are willing to explore, the possibilities are becoming real in truly remarkable ways. We may soon find ourselves in a world where any idea that we can put into words can be visualised in incredible detail.

“I can go anywhere now.”

— Laurence Fuller

As for those who see AI as an existential threat to artists, that those concerns are so widely shared is in itself a reassurance. Baudelaire clearly underestimated the number of people who would continue to love painting as he did, even in the face of photography, as well as the full potential of what photography could be. We do not have to make the same mistake of picking a side in a conflict that need not exist.

“The ability that it has given me is a sort of limitless potential to tell whatever story I want to tell.”

— Laurence Fuller