Will AI Kill Creativity?

This was a question posed in a philosophy forum. Below is my hurried response (slighted edited and extended to make sense as a standalone article).

Will AI kill creativity?

This question is complex, which is itself an astounding fact given that it consists of just four words.

My initial reaction is no, if we mean the brain's capacity to be creative.

Creativity is a biological endowment.

It begins with creative use of language: we can utter novel sentences effortlessly, which is what Descartes found so perplexing. It continues to baffle many bright minds and causes controversy amongst linguists, some of their arguments thrown into doubt by the unexpected creative capacity of GenAI (without the presence of a "creative mind").

The creative faculty of the mind at a cognitive level "above" or "below" language is barely understood and deeply contested as to its nature. Even with new insights from brain imaging, there is still much speculation.

After millennia of philosophical enquiry, there is no agreement about what is art, arguably the most quintessential form of creative expression. (Side note: on this subject, I highly recommend Alva No?'s Strange Tools: Art and Human Nature.)

Within the gamut of ideas of what is art, it is perfectly feasible to argue that what an AI produces is not art, but also that it is art.

And, within some of the descriptions of what is art, it would be easy to suggest that AI has killed creativity. For example, if art includes the emotional interaction between humans in some kind of shared mysterious experience, then it is hard to argue that this is possible with a Synthetic.

There is a body of literature about Conceptual Blending, another biological endowment, namely cognitive synthesis. Even within this rubric, it is perfectly feasible to describe Generative AI as being creative in some sense. It can certainly generate novel outputs based upon conceptual blends. I asked it to generate a rap song about AI in the style of Ni**as in Paris by Ye (see end of article).

The output was novel in the sense of creative use of language, but also novel in the second sense of conceptual blending.

The AI appears to attend to the circumstances (my request) with something appropriate to those circumstances. Had I asked a rapper, lyricist or poet to do the job, the output would have been similar in construction: taking words from one domain (AI) and inserting them into the rhythmic meter of the rap song whilst using Ye/Jay-Z -- i.e. the AI is being creative in conceptual-blending ways, yet far more productively.

(Note: an interesting take on the future of AI in music is this video by Rick Beato. TL;DW = there will be AI artists and human artists. It is unstoppable.)

If we are asking whether or not human creative outputs will decline relative to AI creative outputs, like this faux rap lyrics or the generated images within this article, then this seems almost a certainty because AI tools are incredibly productive.

What is left for the human creator?

If art is only considered as just aesthetic objects, then AI will soon generate far more objects than humans. Within this sense, there is perhaps not much left for those whose skill is producing similar objects (with similar uses) using the techniques that the AI has automated. Even those objects can be conceptually-blended (in their final form): GenAI can create seemingly "novel" expressions of metaphor.

So then we are back to understanding what is art and asking how will its many interpretations emerge and evolve in a post GenAI world.

Different responses will emerge depending upon whether or not the stance is one of pushing back against AI (e.g. placing a higher value on hand-made objects, like pencil drawings, or on authentic experiences like live music) or embracing AI as another artist's tool.

If art is about mastery of skill (e.g. painting a landscape), then new artists will emerge to become masters of AI-assisted crafts. (This happened decades ago with the emergence of Digital Art, a subject I explored when launching a Digital Art PoC product for art.com.)

(Aside: consider the shared etymology of the technique and technology via the root word: tekhnē – knowledge and expertise required to perform a specific craft or trade. I point this out, as I did during my time in Digital Art, because the notion that art exists outside of technology is problematic to begin with, invoking a parochial sense of what is technology, not what is art.)

Anyone can write a prompt and generate an aesthetically useful object, so writing prompts is possibly not what makes a "Generative AI artist" an artist. It's just what makes a regular person an image producer.

So we are back then to the most human of traits: the imagination.

Whoever conjures imaginative ways of using AI will become the new art masters to be held in awe. The complexity of technique is set to sky-rocket with tool-chaining, which only a few will master to the level of audaciousness.

I do recall reading a book of photos generated via chaining of photo-editing apps on mobile. I felt the same sense of awe that one does to any exhibition of mastery: a visceral reaction to the level of human ingenuity involved in conjuring up some of the editing chains. There is no reason to assume that this shared-experience of craft admiration won't apply to AI technique.

Then there is the simple matter of interpretation.

Personally, I enjoy spending time in Midjourney's galleries, exploring the creative milieu of participants, prompt ingenuity and artifice (noting the etymological similarity between artifice to art).

I enjoy re-mixing and riffing another prompt, which feels creative.

And, if I am honest, I have pondered upon some of the images and found that they offer aesthetic invitations that I can accept in order to enter into their aesthetic realms, all of this process not essentially different than interpreting art made without AI.

Of course, there is the possibility that everything is pastiche (because of the common training), but this isn't something new to AI. No doubt, new genres will emerge in due course, given names by taste makers and cultural icons, or iconoclasts.

I am not sure it's the right metaphor to say that in this age of abundance "the bar has been raised", but in some sense it is a truism, but only if we had a limited view of art to begin with, as in it's just the production of aesthetic objects.

And art as a cultural scheme has always had some relationship with audacity. What will be the audacious works in this GenAI era? We await for the creative geniuses to announce themselves or be proclaimed.

Coda: Will AI Dull Personal Creativity?

This was not part of my original response, but I think this is a more interesting question, although quite different, as in will our habitual use of synthetic tools cause our synthesis faculties to atrophy, like how relying upon GPS dulls the sense of geographical and spatial awareness.

As has been demonstrated, habitual thumbing of content on mobile phones might crowd out our access to day-dreaming, with potentially all kinds of deleterious effects, including erosion of creativity and empathy.

This is certainly an aspect that I worry about.

Or what will be the consequence of the ease with which we can quickly attain superficial knowledge of new subjects using ChatGPT? Will this harm cognitive faculties?

Within the rubric of research and bootstrapping into new domains, I have found GenAI to be exceptionally useful. But I don't know to what extent I am making use of sense-making skills previously acquired from years of research where it was my job, not the AI's, to find patterns, form connections and make distillations.

Has the GenAI enhanced existing faculties or given me new ones?

(Note that many of the impressive productivity demos for coding via AI seem to be produced by coders, not dilettantes, or those who have some sense for what code can do. Of course, this sense of code will likely evolve in light of semantic interfaces.)

It feels like there is a potential fragility in relying only upon GenAI to make sense of subjects. But then again, making sense of things often requires acquiring the right framework of understanding and I have found GenAI useful in discovery of different frames. Does this bootstrap or amplify a pre-existing skill built up over many years of research and thinking? I do not know.

I feel like my creativity has been boosted in some tasks, but I have no idea if this is true overall.

When I get in my car and don't use GPS, I often feel lost, even in places I am familiar with. Will this sense of feeling lost emerge when I replace mental maps with synthetic ones? Will it dull my creative faculties? Will it matter?

As this is a philosophical article, I don't have to provide any answers. Indeed, the rush to provide answers about AI with any certainty is a danger of our acceleration towards a left-brain centric world.

But these questions seem worth exploring. And that's the value of philosophy.

THE END

Here's the rap lyrics:

(Verse 1)

Yo, we be chillin' in the lab, it's where we dwell

AI algorithms, man, we programmin' well

From GPT to AlphaGo, we're breakin' every cell

These neural networks got us feelin' like Adele

We trainin' models, makin' history on the daily

We risin' up, ain't no limit to where we sailin'

In the land of AI, humans be prevailin'

Makin' connections like we're writin' in Braille'n

(Chorus)

AI in Paris, yo we takin' over the scene

Machine learning algorithms, you know what I mean?

Ain't no stoppin' us, we got the code so clean

Together we create a future, humanity's dream

(Verse 2)

ChatGPT, the game changer, it's goin' cray

In the city of love, we're moldin' a new way

Data lakes, deep dives, we drownin' in info

These machines be learnin', man, ain't nothin' simple

AI revolutions, we're causin' a ripple

It's a game changer, y'all, we ain't playin' Parcheesi

NLP, computer vision, yeah we got it all

Crackin' codes, complex like a neural sprawl

(Bridge)

We got that OpenAI, it's on the rise

Makin' moves, innovatin', we ain't tellin' lies

In Paris, we're creatin' new allies

AI powered, it's a world we're gonna baptize

(Chorus)

AI in Paris, yo we takin' over the scene

Machine learning algorithms, you know what I mean?

Ain't no stoppin' us, we got the code so clean

Together we create a future, humanity's dream

(Outro)

In the city of lights, we makin' strides

AI changin' lives, it's no surprise

We're in Paris, y'all, where the future lies

Together, hand in hand, we reach for the skies