Creativity in the age of AI

What is a creative process?Someone once told me that in Da Vinci's time, creative professions were taken much more seriously than they are now. Back then, being a designer was a profession like any other, albeit with some preconditions.

Nowadays, in our parts, the "creative profession" seems to have become more of a lifestyle.

Not that every "creative" is necessarily sitting around drinking wine in Antwerp's trendy South district with tattooed arms while posting vague captions on Instagram, but every creative soul has something mysterious and intriguing about them.

This is logical: the creative process is inherently elusive and sometimes even magical. How is it that hours behind your desk, sketchbook, or laptop often yield nothing, while a warm morning shower or a walk in the woods suddenly sparks the best ideas?

In that sense, I understand the mystical haze surrounding creativity. The idea that creativity has been considered typically human for millennia and therefore not automatable explains much of this perception.

No matter how fast technology evolves, a machine could never be truly creative.

And then came AI.

The Nature of Creativity

Just over a year ago, the entire creative sector, especially the world of illustrators and concept artists, was shaken awake by Midjourney, one of the world's first tools that could generate images with just a few lines of text.

Admittedly, the world of AI has taken a few quantum leaps forward since then, but when Midjourney's algorithms brought this image to life in the summer of 2022, I nearly fell off my chair.

Not only because it was technically impressive, but also because I felt that after more than 20 years of searching, I had finally found my medium.

That is to say, for the first time in my life, I could finally visualize the stories that had been wandering around in my head for more than two decades.

"Never thought about picking up a pencil and sketching or taking a creative course?"

I've tried that, several times even, but I just don't have any inherent drawing talent. Besides, there's not a hint of fine motor skills to be found in my body, as evidenced by my handwriting that almost made one of my teachers believe "that cuneiform was back."

The teachers at the art school I attended during high school also empathetically advised me after one semester to put away the brush and follow a different career path.

And rightfully so, because without having walked this life path, I would never have gotten where I am now: between man and machine, trying to answer the question of what creativity really is.

Existential Fear

Where Midjourney and other text-to-image tools were initially met with skepticism, this feeling turned into existential fear within a few months.

Rightfully so. Because it has never happened before that creative professions, typically practiced by a highly educated middle class, have changed so radically in such a short time span.

This brings us immediately to the crux of this article and why I have been tirelessly traversing our sector like some sort of prophet for almost more than a year: artificial intelligence has the potential to take over every human task. And probably faster and better too.

This won't happen overnight, and legislation will likely slow this development to some extent. But that's not the point.

The realization that almost every aspect of our human existence and our culture can be converted into and by an algorithm has profound psychological and philosophical consequences.

At the same time, AI's ability to combine the most diverse styles and references in a way unimaginable to humans shows that creativity still holds many undiscovered secrets

AI = NFT 2.0?

For digital artists, such as certain illustrators and 3D artists, there was never really a platform or revenue model like in the classical art circuit, aside from commercial assignments and the occasional commission.

When NFTs – those digital certificates for art – made their entrance in 2021, I hoped this technology would finally give digital artists the recognition they deserved.

Although it certainly paid off for some artists, the bubble burst quite quickly.

Some tech bros are still looking for the right reason why no one is interested in NFTs anymore, but for me, the cause was clear pretty quickly: viewing art solely as a commodity to be sold as quickly as possible for more money goes completely against the essence of the creative process.

We all know that the classical art world often revolves around money too, but from the moment the creative process becomes transactional, the death of art is near.

It's understandable then that Generative AI is seen as "soulless technology driven by greedy late-stage capitalists," just like NFTs.

Yet generative AI is fundamentally different.

Star Wars, Wes Anderson style

It's true that the people behind Midjourney, Stable Diffusion, and ChatGPT move in the same circles as the founders and investors of the largest NFT projects and crypto exchanges.

Silicon Valley has been the epicenter of innovation for decades. So it's no surprise that generative AI comes from there.

But where NFTs were mainly driven by scammers and money wolves, a very different dynamic reigns in the world of generative AI.

On LinkedIn and X (formerly Twitter), you can't scroll down a pixel without coming across a tech bro trumpeting that "Generative AI will make every creative profession obsolete," but if you dive deeper into certain subreddits on Reddit and chatrooms on Discord, you feel a genuine creative energy.

Look beyond the superficial "Film X or Series X in the style of director Y or illustrator Z" visuals, and you get a glimpse of how a new wave of digital artists is manifesting itself.

But without the work of writers and artists, generative AI wouldn't exist. Also, the fact that most Midjourney users are hobbyists who create purely for pleasure is cold comfort for the many writers and artists whose job security has suddenly been put in jeopardy.

Despite all legislation and "AI Doomerism," the chance is slim that we'll return to a time before AI. So how should we move forward as a sector?

First, I think we need to be honest about the nature of creativity and where human inspiration comes from. The fact that the models of Midjourney, Stable Diffusion, and ChatGPT have been trained without permission on the corpus and work of millions of people is somewhat problematic.

But what's the fundamental difference between an artist drawing inspiration from existing styles & references, and an algorithm doing the same? Generative AI is not a copying machine that exactly replicates works, but rather a kind of "creative calculator" that can discern patterns from styles that we ourselves would never notice.

It's unfortunate, but also inevitable, that this technology has mainly been used in recent months to emulate the exact style of existing artists.

In that respect, I completely understand the anger of artists. The realization that with one prompt you can generate styles that artists have toiled over for years must sting.

Yet I haven't seen a single case of an illustrator being completely replaced by a "Prompt Engineer".

And even if that were to happen, I doubt which companies would be interested in an imitation.

If we set aside the philosophical for a moment and look at the real impact, I don't think many companies will risk blatantly copying a particular artist. That was already the case before AI, and it won't change in the short term.

What I've mainly learned in the past ten years in the creative sector is that most brands are afraid of controversy and lawsuits.

Except for companies like Zara and Shein, but they didn't need AI to behave unethically for decades.

Artist vs Curator

Exact copies may be banned, but what about so-called AI artists who combine 20% Hergé, 25% Mœbius, 15% Ledroit and 40% Michaël Borremans?

This brings us back to the essence of this discourse: what is a creative process? Where does inspiration stop and plagiarism begin? And who determines what art is?

By the way, if we expect 100% transparency from an AI-generated artwork, are we going to demand the same for 'human art'?

Some artists claim that art without creating yourself has little value. Even though I realize all too well that not everyone has the skills or physical capabilities to draw or paint, I tend to agree.

I don't really see myself as an artist either. Rather as a kind of creative shepherd cum curator.

When I bring the world of Starhaven to life by combining the style of video game engines with the Romantic painting style, I don't pretend to do the same as a 3D artist or painter.

But I do create.

The same goes for the Arboristiae, or organic quantum computers that form the bridge between the world of algorithms and ours.

In addition to combining totally different styles, I also use the power of algorithms to blend Celtic and the Amazigh language from Morocco to come up with new names and terms.

Is that language grammatically correct? No, but for me, it's something different than the umpteenth science fiction artifact bearing a Roman or Greek name.

The large language models behind ChatGPT are primarily prediction machines that predict what should follow based on your prompt, which sometimes makes them predictable.

But both ChatGPT and Midjourney are capable of so much more. Everything depends on how broad your frame of reference is and how you communicate with it.

If the untrained hobbyist can already generate impressive things, imagine what an artist who knows everything about visual language can do with it.

That's not to say that all art should be algorithmized by definition, but I do think this approach should encourage artists to think differently about technology and creativity.

Anthropomorphism

Despite our tendency to anthropomorphize things we don't understand, AI (still) doesn't have consciousness, let alone a creative soul. Nevertheless, I have already seen AI generate things that genuinely move me.

I often compare it to the "suspension of disbelief" we experience when watching a film or reading a comic; if the storytelling and worldbuilding are coherent and the dialogues touch us, it doesn't matter to our brain and heart that it's fiction. We choose to suspend our disbelief.

What I do wonder is whether we can do this with films or series with synthetic actors. I can't articulate it perfectly, but a 'real' actor identifying with a role has a certain value for me.

Even in films like Avatar, where a lot of CGI was used, you feel the humanity through the digital layers, because it's acted by real actors.

Will a film that is 100% synthetic but realistic evoke the same feeling? Probably not for me, but I wonder how generations who have never known a world without AI will experience this.

This immediately raises a relevant side question: in which professions do we find it acceptable for them to be taken over by AI?

Everyone is rightly concerned about the fate of writers and actors in Hollywood who will probably still be in a battle with the big shots of Disney, Warner Bros, Netflix, ... while writing this.

But what about the VFX artists?

In the advertising world, it's often said that an advertising agency makes the client look good, the producer makes the advertising agency look good, and the actor makes the producer look good. But people often forget that the VFX studio makes everyone look good.

And yet there was hardly any solidarity a while back when the VFX studio behind "The Life of Pi" went bankrupt, shortly after winning an Oscar. The workload was unreasonably high.

Looking at our own sector: will we only feel bad about job losses for illustrators and writers, or also for the hundreds of thousands of creatives from India or Sri Lanka who lost their jobs due to Photoshop's "Generative Fill"?

I don't know any designer who looks forward to removing backgrounds in Photoshop. That's why websites like Fiverr exist: to outsource that kind of work.

Midjourney was immediately vilified, while Adobe's Generative Fill, which allows you to expand an existing photo or add details with AI, was generally accepted.

I realize there's a difference between making a complete illustration and digitally editing backgrounds, but the question remains relevant.

Consumption vs Interaction vs Creation

We may never be able to answer what the true nature of creativity is. It's also difficult to predict which jobs will definitively disappear.

What we can do to broaden culture in the broadest sense of the word is to reconsider the traditional boundaries within pop culture and entertainment.

Roughly speaking, we can divide our relationship with culture into three categories: consumption, interaction, and creation.

There are plenty of examples of cultural forms that combine them, just look at video games like Minecraft. With films, books, and comic books, on the other hand, it's mainly about consumption.

I'm not going to watch Dune ten times in Imax to interact with it or to transform myself into a sandworm, but to be engulfed by the storytelling and worldbuilding.

With World of Warcraft, it's long been not about the story for me anymore. I play it to be able to fly on dragons and use spells while trying to save an anonymous fellow player from Sweden from certain death.

What makes AI so revolutionary is that it can blur the lines between consumption, interaction, and creation.

If building a virtual world becomes as easy as walking around in it, the possibilities suddenly become much greater.

The number of hours I put into World of Warcraft or video games in general has decreased considerably since the rise of Generative AI.

Bringing to life Starhaven's Heralds, whose task it is to help machines dream so that they better understand our world, gives me more artistic catharsis than any video game.

New Forms of Expression

The point is that companies like Midjourney are not primarily aimed at replacing traditional media such as comic books. Their explicit goal is to develop new media and broaden human imagination.

I would normally classify this kind of statement as "boasting from yet another tech bro", but the fact that Midjourney is self-sufficient and not driven by the typical start-up urge to grow and sell quickly makes it more credible.

The graph below from the founders of Midjourney will undoubtedly strike fear into the heart of many classically trained artists, but it shows that Midjourney is working with a long-term vision.

And the beauty is that in their vision, comic books, analog films, and works by painters can coexist. In fact, the rapid technological progress seems to stimulate a revival of crafts and 'tangible' art and design.

My youth consisted of Star Wars and Marvel on one hand, and Buck Danny, Requiem: The Vampire Knight and Thorgal on the other. With my iPad, I can relive these stories in ultra-high resolution at any moment, but I consistently choose the battered copies from 20 years ago. Tears, creases, and fossilized Prince cookie remnants included.

Will there be people trying to sell superficial and plagiarism-drawn AI comic books? Absolutely, but the real comic lover will always choose the original.

AI will irrevocably change our sector. That's inevitable. A number of artists, styles, and media will probably not survive.

But if AI contributes to the downfall of yet another soulless blockbuster with a budget of hundreds of millions, I'll gladly raise a glass of champagne.

While the big Hollywood studios refuse to pay actors, writers, and VFX artists fairly, you see independent studios like A24 flourishing.

Even though capitalism is always lurking around the corner, the big entertainment moguls seem to realize that they can't keep ramming pop culture junk food down our throats.

Victory for the creative proletariat then? Not quite.

The Style Singularity

AI has democratized quite a few creative professions in a short time. In addition, the technology also shows that we're not at the end of our Latin when it comes to storytelling and mediums.

However, the disadvantage of an algorithm that can combine innumerable styles is that it can combine innumerable styles.

Where it was already difficult to keep up with the latest trends due to globalization and the rise of TikTok, with AI we're probably heading towards a "style singularity".

In other words, in the coming years, more styles will probably be created than have ever existed in human history.

But just like quantum particles around a black hole may only exist for a fraction of a second, artificial intelligence produces styles and art movements daily that disappear again within a week.

This dynamic is inherent to AI art. It has become so easy to create such stimulating visuals that their value inevitably decreases.

As a result, the connection with the artist becomes secondary, while this is a crucial element in traditional art.

Anyway, the observation that AI is capable of delivering clever and stimulating ideas without an inner artistic quest or dialogue with the environment is a context that no artist can ignore.

I'm not saying this from an "adapt or get out" mentality, but from the sincere belief that traditional artists and designers will be more important than ever. Just as my father and grandfather, both sea captains, navigated the seas with stars and charts, I see us as digital guides in a sea of algorithmic creativity.

Before AI, there was only budget, time, and creative bandwidth for a handful of illustrations, texts, or scripts. Now there is an abundance of choices, and unfortunately, that infinity can be paralyzing for many.

The challenge will lie in finding the balance between daring to let go and daring to say 'No' to technology. We must learn when to loosen the reins and let the machine do its magic, and when to intervene to put our own unique stamp on our work.

In this new reality, where technology and creativity merge, it's essential to understand that AI is not a replacement, but rather an extension of our own creativity. It's a tool, just like a pencil or a camera, which in the right hands can lead to unseen results.

At the same time, it's also completely legitimate and valuable for an artist to go completely analog and ban technology from the creative process. Handmade illustrations have an undeniable value that cannot and need not be replicated, even by the most advanced algorithms.

Although AI still shows no signs of consciousness, it has shown us that it's capable of creating works that can move us. This raises fundamental questions about the nature of creativity and inspiration. Is creativity solely human, or is it something universal, something that can be shared and expanded by technology?

I'm inclined to say yes and no longer see a text or artistic work as the endpoint of a long process, but rather as the starting point of something new and grander.