Neuroscience Of

View Original

Generative AI Separates The Art from The Artist. Humanism isn’t Happy


This is the second of a multi-part series on the influence of Generative AI on music and human creativity. If you haven’t already, you read the first piece here.


Terence began life as a slave. Born in Carthage in 195 BC, he was brought to Rome as a servant for a Roman Senator. Over the years, however, he was given an education, and ultimately, liberation. 



Once a free man, he blossomed into the most beloved playwright of his era, one whose art survived over 2 millennia, and whose ideas continue to resonate today. His plays, often accompanied by music, became standard curriculum throughout the Roman Empire, and later, throughout European theater education. 



He is credited with being one of the key pioneers of humanism - a broad philosophical movement, encompassing centuries of thinkers and artists, oriented around an appreciation of humanity, an ethics of compassion, and an exploration of human connection. For over 2,000 years, Terence’s work has embodied the humanist spirit



His work has found deep resonance with people across disparate eras, cultures, and geographies, inspiring thinkers including Erasmus, Martin Luther, Shakespeare, Thomas Jefferson, and Maya Angelou. 



Art as a Medium for Human Connection



Perhaps more than anyone else, Terence’s work attests to the incredible power of art as a medium of human connection. This capacity is beautifully distilled in one of his most famous lines, “Homo sum, humani nihil a me alienum puto” which translated to “I am human, and I consider nothing that is human alien to me”. 



His enduring legacy is a reminder that audiences connect not just to the art, but to the humanity of the artist. When you experience a work of art, you can feel the humanity of the artist staring back at you. What happens though, when there’s just the art itself? Stimuli with no trace of humanity for the audience to connect to? 



For most of human history, this question would have been absurd and meaningless. The art and the artist are inextricably linked. Increasingly, however, a new wave of emerging technologies is pulling the two apart. Generative AI now presents us with the art and music that is derived independently of the human condition. And in turn, in the words of Terence, an art that is truly alien. 



What will Generative AI mean for art, and this capacity to connect? Let’s dive in by narrowing in on music, the newest form of art to see AI invade it’s creative process.   



Suno AI gives birth to the Soul of The Machine



There’s been a long history of bad AI music. But as we’ve seen, since 2023, AI-Music has reached a new level. Perhaps not yet great, but still, not safely dismissable. 



2023’s Heart on My Sleeve - a slow trance rap featuring the AI voices of Drake and The Weeknd, was one of the first AI-derived tracks that fans within the genre genuinely enjoyed. Since then, the field of AI Music has only accelerated. The biggest innovation in AI music? Democratizing musical creation. 



Compared to the other creations of Generative AI, which now includes sophisticated text (e.g. ChatGPT), images (e.g. DALL-E), and video (e.g. Sora), music has trailed the field. In early 2024, however, one company in particular has taken AI music a major leap forward: Suno AI



You can think of Suno AI as the ChatGPT for music. Like all LLMs, Suno is trained on an untold amount of data, and uses a series of complex transformer algorithms to produce novel output based on this training data. You describe what you want to hear, and Suno births it into being. 


The text prompt, “solo acoustic Mississippi Delta blues about a sad AI”, for example, resulted in the song, Soul of The Machine, a stunningly realistic minute-long song. It’s an enjoyable, realistic version of what you’d expect, from a “solo acoustic Mississippi Delta blues about a sad AI.” And it sounds like the kind of creation that a soulful blues artist from the Mississippi Delta would themselves create, if they were so inclined to write a song about a sad AI.  



Suno AI represents a massive leap forward in AI-derived art. With Heart on My Sleeve, legitimate questions existed about AI’s role in the actual composition of the song, above and beyond AI’s role in production and vocal mimicry. But here, these questions are extinguished. AI is the unambiguous creative engine. It creates music with no identifiable residue of humanity.



Does Great Music Require the Human Experience?



In part 1, we looked at the question of whether AI can make great music. On the one hand, this is a question that can only be answered in time. If it can, it will, and if it can't, it won't. 



But there's another perspective on this question. In this view, the answer is already there: No amount of time + technology could ever add up to produce great music. The idea of an AI creating great music is absurd - not merely in practice, but impossible in principle. In this view, asking if a non-human can make great music is a bit like asking whether you can make coffee without using coffee beans. In the creation of music, the human condition is the most important, and the most required ingredient.



This argument has intuitive appeal. Nearly every culture that has ever been studied has its own music, and we're the only species that creates and enjoys this kind of art. From our earliest years as nomadic tribespeople, gathering around the campfire over 200,000 years ago, music has always been with us



As the author and musician Clemency Burton-Hill describes, “We are a music-making species — always have been, always will be — and music’s capacity to explore, express and address what it is to be human remains one of our greatest gifts…It is an impulse that is still fundamental to who we are.”



Few can disagree that music is amongst our most treasured abilities. But are we the exclusive holders of this craft, now and forever? 



Music Without the Humanity in The Creative Process



Many say yes. A prominent spokesperson for this perspective is The Australian folk singer Nick Cave. In 2023, his fans presented lyrics created by ChatGPT, “in the style of Nick Cave”. Impressively, the lyrics captured some of Cave’s signature humor and included his dark religious imagery. 



Cave, though, wasn’t impressed. In his words, the song “sucked.” Of course, opinions can differ about the quality of the lyrics, but when we examine Cave’s criticism, we can quickly see that it's not an argument about the musical product itself, but what he thinks about the mechanism of its creation: That music can only be great because it is made by a human.



As he elaborated to the BBC, the “creative” output of generative AI can only mimic that of humans. "It could perhaps in time create a song that is, on the surface, indistinguishable from an original, but it will always be a replication, a kind of burlesque..Songs arise out of suffering, by which I mean they are predicated upon the complex, internal human struggle of creation and, well, as far as I know, algorithms don't feel. Data doesn't suffer.”



Whether we agree in full with Cave’s outrage or not, he touches on a key critique of AI-generated art: the source of the creative process is fundamentally flawed. AI is too superficial, and too unhuman to ever create art. It’s a mimicking machine, not a source of artistic wonder.. As Cave states, “ChatGPT has no inner being, it has been nowhere, it has endured nothing.”



Here, AI can’t ever create great music since it isn’t born of human nature. What should we make of this argument? 



How Generative AI Separates the Art from The Artist



Musical perception, like the perception and appreciation of art, more generally, is complex. What we consider to be “great music” isn’t as simple as what it makes us feel the first time we listen to it. Instead, it’s about what we come to think about it, and how to speaks to us. In a 2024 paper, Duke scientist Lucas Bellaiche and his co-authors make this distinction explicit, dividing it between “art as (1) a purely physical stimulus and (2) a deeper communicative medium of the human experience.”


In other words, there’s a level of enjoyment that comes from the art itself in how it impacts our senses. And there’s a deeper level of meaning that it can convey about the human condition - the kind of connection that makes Terence’s work, over two millennia old, feel personal and resonant. 


Generative AI is fully capable of teasing apart the art and the artist. Can we come to fully enjoy music, and art more generally, which is devoid of humanity? To understand this, we need to better understand the psychology behind this second mechanism: art as a deeper medium for communicating the human experience. This is where we turn next.


Photo by Billetto Editorial via UnSplash


This is a multi-part series on The Psychology of Generative AI in Art and Music. Next, we explore the psychology of essentialism and how it influences our perception of AI art. Be the first to get the the next piece by signing up here (for free)



References for Generative AI, Musical Creativity, and Humanism

Alexander, A. (2024). “Heart on My Sleeve”: An AI-Created Hit Song Mimicking Drake and The Weeknd Goes Viral. SAGE Publications: SAGE Business Cases Originals.

Franko, G. F. (2013). Terence and the Traditions of Roman New Comedy. A companion to Terence, 33-51.

Hiatt, B. (March, 2024) A ChatGPT for Music Is Here. Inside Suno, the Startup Changing Everything, Rolling Stone

Savage, M. (Jan, 2023) Nick Cave says ChatGPT's AI attempt to write Nick Cave lyrics 'sucks', BBC News

Verma, S (2021). "Artificial Intelligence and Music: History and the Future Perceptive". International Journal of Applied Research. 7 (2): 272–275