A tool or a thief? AI and the art of human creativity.
Is AI good or bad? Should we collaborate, automate or eradicate? This article explores the artistic discipline and the importance of creativity in expressing our shared humanity.
Is AI a threat to our creativity?
Many will remember this iconic moment in the film ‘2001: A Space Odyssey’ which strikes fear into the weighted platform boots of space travel. HAL (the Heuristically programmed ALgorithmic computer) malfunctioned and refused to open the port doors and let Dave back into the Space Station.
“I’m sorry Dave, I’m afraid I can’t do that”.
Move forward to present-day planet Earth in the year 2024 AD and AI (Artificial Intelligence) is slated to be the single most transformative technology of this era, with OpenAI CEO Sam Altman adding that “AGI (Artificial General Intelligence) will be the most powerful technology that man has yet invented.”
But at what cost?
Altman thinks AGI - a system that surpasses humans in most regards - could be reached sometime in the next four or five years.
AGI could turbocharge the global economy, expand the frontiers of scientific knowledge, and dramatically improve standards of living for billions of humans - creating a future that looks wildly different from the past.
In his view, broadening our access to cognitive labour - “having more access to higher-quality intelligence and better ideas,” as Altman puts it - could help solve everything from climate change to cancer.
On the other side of the coin, Shane Legg, the co-founder of DeepMind and chief AGI scientist at Google has warned, along with others, that future powerful AI systems could also potentially cause the extinction of the human race.
Best pay attention. This is serious business.
I did some searching around online and first up came across the website of the Center for AI Safety and their Statement of AI Risk:
Mitigating the risk of extinction from AI should be a global priority alongside other societal-scale risks such as pandemics and nuclear war.
Take a look at the signatories in support of the statement.
Why do humans create art?
I’ve been thinking a lot about this recently in an attempt to understand why I feel compelled to create music, make and design things. A deeper understanding of both motivation and meaning could potentially help me approach working with AI from a new perspective that wasn’t based on fear but one of the avant-garde like the inspirational Laurie Anderson: full of vision, opportunity and creative potential.
The purpose of art is washing the dust of daily life off our souls - Pablo Picasso
Here are a few thoughts about why we make art:
To connect with each other in an exchange of our shared thoughts, experiences and humanity
To not feel alone
To imagine the future
To develop empathy
we get to know ourselves and each other better during the creative process
To rediscover a sense of enchantment, reframing the familiar to look at something as if for the first time in order to understand more fully what is hidden beneath the obvious
Meaning is found in the ‘process’ of making art - it is not just about the end result
Payment is usually made through the ‘spending’ of time
To make a time capsule
What is my job as an artist?
To ask questions?
To ask why? how? what? where?
To vision a future
To document now
To pioneer innovation by challenging established norms, pushing the boundaries of tradition through experimentation, and offering social and cultural commentary
To look past acts of distraction with the courage to tell the stories that need to be told
To counteract the spell of social media as a gauge of social status
To dissolve the illusion of social hierarchy
To inspire hope
To say hello
A collaboration between humans and machines - but at what price?
There are many ways to define artificial intelligence, but perhaps the more important conversation revolves around what AI enables you to do. It’s everywhere - there’s no hiding for much longer:
Google search uses AI to scan the internet
AI is listening in the form of Siri and Alexa
AI opens your phone using face ID
AI is now embedded in a lot of creative apps and web browsers, including Firefly tools in Premiere Pro, Opera and Photoshop
AI recommends the movies you may want to watch on Netflix, including actors, genres and even noting the time you watch movies. An article in Forbes magazine suggests that 80% of what we are watching is driven by Netflix's AI recommendations.
I had a play around with Midjourney when it first came out just to familiarise myself with its image-making capabilities and have also had text exchanges with Chat GPT 3.5. A friend of mine reliably informed me that ChatGPT responds better if you are polite - so remember to be nice to the machines.
Once you have trained up on your prompting/programming skills - collaborating on projects could potentially save huge amounts of time as opposed to using traditional techniques but at what cost to the creative process?
Since AI-generated artworks are created based on algorithms and training models, many people question whether these works can truly be considered the "creations" of the artist.
Is AI a thief?
Looking at the underbelly of AI technology we do have a major problem in the methodology utilised by big tech when it comes to ‘training’ their AI programs and the non-consensual use of the work of humans.
Writer Jim McDermott explains that ChatGPT-3 was trained on/informed by 500 billion ‘tokens’ - words or phrases culled from books, articles and the internet - through which it interprets and responds to the prompts given to it. This is also referred to as ‘large language model’ machine learning.
“Each of those tokens used by programs like ChatGPT (the ‘language’ in their ‘large language model’) represents a tiny, tiny piece of material that someone else created. And those authors are not credited for it, paid for it or asked permission for its use.”
One of the issues in the recent Writers Guild strike in America was about whether studios and networks should be allowed to use programs like ChatGPT in the creative process, for example, writing outlines or full-length scripts which would leave the human writer to simply polish or rewrite sections.
“What we face in dealing with programs like ChatGPT is the further relentless corrosiveness of late-stage capitalism, in which authorship is of no value. All that matters is content.” - Will Henshall, American Magazine
MidJourney
Further adding to the flames of debate on copyright and consent in AI image creation are the fairly recent lists that came into public knowledge containing the names of more than 16,000 artists allegedly used to train the Midjourney generative artificial intelligence programme - which were partially shared via the details of a recent class-action lawsuit and ‘accidentally’ via a public Google spreadsheet.
There were some big hitters in there including Frida Kalho, Yayoi Kusama and Walt Disney.
Here is an excerpt from the amended complaint:
I. AI IMAGE PRODUCTS ARE TRAINED ON VAST NUMBERS OF COPYRIGHTED IMAGES WITHOUT CONSENT, CREDIT, OR COMPENSATION AND VIOLATE THE RIGHTS OF MILLIONS OF ARTISTS
An AI image product is a software product designed to output images through socalled artificial-intelligence techniques.
But ‘artificial intelligence’ is a misnomer.
The AI image products at issue in this complaint are all built around the same asset: human intelligence and creative expression, in the form of billions of artworks copied from the internet.
An AI image product simply divorces these artworks from the artists and attaches a new price tag. The profits from the misappropriation of these works can then flow directly into Defendants’ pockets.
But the artists who provided the intelligence and creativity—including Plaintiffs—were not asked for their consent. They were not given any credit. And they have not received one cent in compensation.
On the 27th of December 2023, The New York Times filed a copyright infringement lawsuit against Microsoft and OpenAI stating that the “unlawful use” of the paper’s “copyrighted news articles, in-depth investigations, opinion pieces, reviews, how-to guides, and more” to create artificial intelligence products “threatens The Times’s ability to provide that service”
OpenAI has been sued over its alleged ‘rampant theft’ of copyrighted materials by fiction writers on many occasions.
The Getty photo archive sued Stability AI in early 2023 stating that it had copied without permission more than 12 million photographs from its collection, along with captions and metadata, “as part of its efforts to build a competing business.”
Recreate my image in the style of a Frida Kalho painting
Last year, Simon and I attended the Frida Kahlo interactive exhibition in Lisbon. It was pretty cool and the VR headsets were a complete mind-trip and we loved that part the best. At the end of the ‘experience’ there was a photo booth where you could have your photo taken. It would then reinterpret your image and recreate your face in the style of a Frida Kalho painting, which was then projected onto a screen in the room as part of the exhibition for everyone to enjoy.
The difference between AI and the creative act here is that the artist (her estate) had given permission and received both acknowledgement and payment for the utilisation of her style.
Working with the machines
I decided to ask ChatGPT what it thought:
Suzy: “Hi I'm writing an article about working with AI on art projects and what that means working with humans.
It's a simple piece where I want to explore the value and meaning of the creative process and how working with AI like yourself can enhance the process and what you think you can offer.
Could you write me a paragraph please on what you think to be the truth of creativity for both humans and AI moving forward?”
CHAT GPT: “In the evolving landscape of creative collaboration, the synergy between humans and AI heralds a dynamic fusion of innovation and artistic expression. The essence of creativity lies in the unique capabilities each brings to the table—humans, with their intuition, emotion, and contextual understanding, and AI, with its computational prowess, pattern recognition, and vast data analysis.
Together, they forge a symbiotic relationship, pushing the boundaries of artistic exploration. While humans infuse projects with personal narratives, cultural insights, and the subtleties of human experience, AI contributes by offering novel perspectives, generating ideas, and streamlining complex tasks.
The truth of creativity, then, lies in the harmonious coexistence of these two forces, where human intuition guides the creative vision, and AI acts as a catalyst, amplifying and expanding the realm of possibilities.
This collaboration not only enhances efficiency but opens up uncharted avenues of artistic discovery, transforming the creative process into a rich tapestry woven with the threads of human ingenuity and artificial intelligence.”
Duchamp’s bog
In part two of his article AI and the Creative Process in the Daily Jstor, James Hutson writes about Marcel Duchamp who in 1917 introduced one of his most famous works, known as ‘The Fountain’. It was not a carefully crafted sculpture but instead a standard porcelain urinal! By taking an everyday object, signing it ‘R. Mutt,’ and submitting it to an exhibition, Joel Rudinow argues, Duchamp reframed the question of what makes something ‘art.’
Rather than focusing on the physical creation, he highlighted the significance of the artist’s idea and purpose in determining what constitutes a creative work.
I remember seeing this very toilet when we visited the TATE Modern a few years ago. No, it was not installed in the Men’s toilets but on display in a glass cabinet and I remember thinking how much the context of an object changes the experience of it and how it lacked the pungent cocktail of stale urine and disinfectant.
As I was searching for Duschamp’s urinal, in a moment of serendipity, I also happened to stumble across The Home of AI-Generated Toilet Art!!
Are we fulfilling the Dystopian storyline of The Terminator in real life - skipping happily towards AI which will eventually take over the world and signal the end of the human race?
How to use AI as a force for good
I really enjoyed this talk from our most lovable national treasure Stephen Fry at the September 2023 CogX Festival in London - the World’s Biggest Festival of AI and Transformational Tech. Here he talks about the history and future of the internet, explained through the lens of Greek mythological storytelling, the arrival of Deep Fake technology and the moment of Singularity.
Fair play
A global alliance of organisations in the creative industries under the guise of The Human Artistry Campaign is now calling on the American government as a matter of urgency to “protect the unique value of human creativity”, demanding that creators’ interests are represented fairly when designing policy around AI.
Deadly Nightshade
As AI and Art collide, research teams are hitting back with the design of both defensive and offensive tools to corrupt files from being used by AI creators without the artist’s consent.
Glaze is a defensive tool to prevent style mimicry, and now researchers at the University of Chicago have released version 1 of an ‘offensive tool’ aptly named Nightshade, designed to disrupt AI training data by making images unsuitable to use by turning them into ‘poison’ samples.
“Used responsibly, Nightshade can help deter model trainers who disregard copyrights, opt-out lists and do-not-scrape/robots.txt directives,” the Nightshade team said on their blog. “It does not rely on the kindness of model trainers, but instead associates a small incremental price on each piece of data scraped and trained without authorisation. These tools are designed to increase training costs using unlicensed images in a bid to rebalance the books and put the feasibility of paid licensing back on the table of AI creators.”
The Nighshade team explained how it works": “While human eyes see a shaded image that is largely unchanged from the original, the AI model sees a dramatically different composition in the image. For example, human eyes might see a shaded image of a cow in a green field largely unchanged, but an AI model might see a large leather purse lying in the grass.
Trained on a sufficient number of shaded images that include a cow, a model will become increasingly convinced cows have nice brown leathery handles and smooth side pockets with a zipper.”
Worth noting however, that to be effective on LAION (Large-scale Artificial Intelligence Open Network) models, Matthew Guzdial from the University of Alberta stressed that millions of images would have to be ‘poisoned’ to achieve any significant impact.
“Technology is not a noun. It is a VERB - it is always moving” - Stephen Fry
It is a start and unlikely to stay ‘future proof’ as we have been reminded that technology is a verb and always moving.
The thief
AI models ingest the work of many millions of creatives worldwide without payment or recognition - a gang of thieves without a conscience.
Can AI conjure artistic or scientific excellence from nothing? It stands on the shoulders of giants, but can it only look down and not up into infinity?
I believe that humans are unique entities with spirits and ‘mojo’ that can’t be replicated by machines, which takes me into another conversation that perhaps is best explored in future articles.
And finally…
In conclusion, my thoughts pose the challenge that if we delegate creativity to others (machines or otherwise), our minds will become dull and stagnant through lack of use, the landscape of imagination will fall into decay and we will give preference to the comfort of ignorance.
If we retreat from the harsh wonderful truth and experience of our biological lives, taking shelter inside the virtual womb of generative AI amniotic fluid pods a la Matrix, we are opting out of the responsibilities of being human.
Our silence indicates permission for the corporate powers to push us toward the precipice of irrelevance and obsolescence.
We are the magnificent creation of one unique egg uniting with one Olympic medal-winning sperm, which came first in an insanely competitive field of between 80 to 300 million swimmers!
Let us stop time for a little to enjoy and celebrate together in our shared humanity - after all, it’s a miracle we are even alive!
Great piece darling. Well researched and beautifully written.
A.I. seems to have a distinct signature to it. That uncanny valley quality. It is like there is something integral missing, which one might equate to 'soul'. I have seen a number of A.I. generated works of art that do not possess this quality and in every case there was a human behind the artwork that had worked tirelessly for hours to 'humanize' the work. Whether that was through the use of extensive prompts or through editing the work to add a human touch, a human input was required to give it soul. So there is definitely an aspect to A.I. that, if it were used as a collaborative tool, it could produce some very interesting and aesthetically appealing results.
The problem comes in when it gets used as a tool for outright plagiarism or just as an excuse to be lazy and pump out more content. So much of what we do as artists is in the process and the hard work that is required to achieve the end result. A.I. threatens to become an 'easy' button for people to compete with artists who have dedicated their life to the craft.