AI, Personhood and Creativity
Human beings are creative – AI is not:
I wrote recently about how man is not a machine. Yes, we can be thankful for labour-saving machines and devices and enjoy how much easier life has become because of various scientific and technological breakthroughs. But try as they might, machines will never replace people.
I mean, they might well replace us with malicious intent, but they will never be able to be just like real human beings. This is true in so many areas, including human creativity. The arts is one area that of course comes to mind, be it things like music, poetry, literature, or painting.
Machines and art
Machines do what they are told to do, or designed to do, or programmed to do. Genuine creativity is the stuff of persons, not machines. So a machine ‘doing art’ might be an oxymoron. One recent book on AI that speaks to this is Non-Computable You: What You Do That Artificial Intelligence Never Will by Robert Marks (Discovery Institute, 2022). I have quoted from it previously: https://billmuehlenberg.com/2026/08/30/are-humans-simply-machines/
Chapters 3 and 4 of the book discuss putting AI to the test when it comes to creativity, in terms of writing as well as in terms of art such as music and painting. Says Marks:
“AI claims of creativity quickly fade when examined more closely. AI trained on examples can only mimic and interpolate among accumulated inputs. But creativity requires discarding dogma resident in the database. Creativity requires extrapolation. Creativity requires transcending boundaries. Creativity comes from ‘thinking outside the box’.” (p. 57)
On this matter of ‘thinking’ new thoughts, Marks says this:
AI’s inability to think outside the box was noted by MIT’s Patrick D. Wall as far back as the 1960s. He said:
“I don’t believe that any of the machines that we know today can think. I have a basic question. Do these machines produce anything really new? When you consider the great new ideas produced by men like Newton, and Darwin, and Galileo, you’ll find that initially they had to throw away the old rules that they’d been brought up with. Machines do what they’ve been told to do. They obey the rules that have been fed into them by man. And we know of no machines at present that have means of overcoming this limitation.”
We still don’t. Humans, however, have the mysterious ability to at times transcend received knowledge, and the boundaries of established belief – to look beyond. (pp. 57-58)
Consider just one such area: AI and music. Writes Marks:
Popular music’s popularity is due in large part not to the sophistication of the music, but to catchy melodies, melody and rhythm variations, fun lyrics, musical hooks, blues improvisation around the pentatonic blues scale, and – of course – the performer’s singing skills, stage presence, and emotionally charged performance, including such things as note bending by stretching guitar strings to convey emotion. (p. 80)
He looks at the music of the Rolling Stones as an example, and then says his:
Music’s appeal is highly correlated between the emotional connection between the artist and the audience. In popular music, showmanship is mandatory. When bending a blues note string on his guitar, Stevie Ray Vaughan’s guitar face adds a lot to the song’s presentation and the connection between the musician and the audience. Emotional performances enhance enjoyment of the music.
AI, however, can’t do emotion. Recent studies have shown that while AI can do well when it comes to cognitive-oriented advertising (ads appealing to consumer’s minds), AI is ineffective when it comes to emotional appeals. “Human rather than AI input is needed for creating emotion-oriented advertisements,” the authors of the studies conclude.
But emotion is what makes music work – the emotion conveyed by the performers, the emotion the music evokes, the emotional connection between audience and artist. Much of that emotion and connection rely on personality. And as software architect Brendan Dixon notes, “This is the blind spot of AI creativity. There’s no one home. There’s no ‘personality’ behind the ‘creation.’” (pp. 81-82)
Marks goes on to look at jazz. My cyberspace friend Douglas Groothuis is both a big jazz fan, and an AI sceptic. So I recently sent him the following section from the Marks’ book:
Jazz poses an insurmountable problem for AI. ‘You cannot reduce jazz to mere repetition or formula,’ computer architect and jazz enthusiast Brendan Dixon says. ‘AI can’t do jazz because spontaneity is at jazz’s core.’
Jazz musician and music critic Ted Gioia is on the same page, saying, “More an attitude than a technique, the element of spontaneity in the music rebels against codification and museum-like canonization.” He says: “Some years ago I worked with an expert in computer analysis of rhythms, and together we tried to understand what was actually happening to the best in music that possessed a strong sense of swing. What we learned was that especially exciting performances tended to break the rules.” Breaking the rules is going outside the box – a necessity for creativity.
As we saw in the previous chapter, AI can’t break the rules. And following rules too carefully in jazz is fatal. Gioia says that at a “rudimentary level of performance, the musicians tend to rely repeatedly on a small number of rhythmic patterns in their phrases. Even if the notes they play are different, the rhythmic structures of the phrases are often identical. Such improvisors might sound convincing for a single chorus, but if the solo goes on long enough, even novice listeners will perceive an inescapable monotony in the proceedings.”…
Like much of AI, machine-generated music can be used as a tool by songwriters. A smorgasbord of AI-generated hooks and tunes can be mined and enhanced by the composer interacting with the AI. But the creativity involved belongs to the programmers and to the composers whose works were fed into the AI as training. It does not belong to the machine itself. Nor can a machine meaningfully participate in that most human of artistic enjoyments, the live performance, with its myriad human connections. (pp. 83-84)
Machines and writing
While I do quite like music, the matter of writing especially concerns me, since I am a writer. I just came upon this rather alarming quote that comes from a piece penned a few days ago:
A study by the University of Maryland found that more than 9% of all news content found in newspapers in the United States had text generated by AI. Different studies found that up to 53% of all books published in 2025 found on Amazon had some level of AI-involvement and over one-third of all publishing online showed signs of AI authorship since the release of ChatGPT. https://dailytrojan.com/2026/08/27/artificial-intelligence-is-killing-the-creative/
That sounds quite problematic to me. While some writers thrive on AI and depend on it heavily, not all of us do. I happen to give it a wide berth. I have one friend who keeps insisting that I give ChatGPT a go. I keep saying I might – but I never do! Perhaps one day, but not just yet…
Let me share two brief quotes from my cyber friend Doug. Last year he posted this on the social media: “I have never used a ghost writer nor AI for writing in my name. I never will. I believe in authentic authorship under God Almighty.”
And just the other day he posted this: “Declaration to the World: No generative AI has been used in any of my books, booklets, articles, reviews, letters to the editor, facebook posts, syllabi, or cards. So there. I am still an author.” In between these two Dougisms, he even drafted an essay titled “We Refuse it: A Manifesto on Generative AI Writing”. I discussed it here: https://billmuehlenberg.com/2026/07/17/christian-ministry-and-ai/
I am with him 100 per cent. I will not use AI to write anything, and I can tend to spot AI-generated ‘writing’ a mile away when others are relying on it. I don’t like it. And I need to appeal to something that Robert Marks reminds us of. He shares a quote from Orwell’s dystopian novel 1984. It had actually forecast a world where AI writes novels:
Julia was twenty-six years old… and she worked, as he had guessed, on the novel-writing machines in the Fiction Department. She enjoyed her work, which consisted chiefly in running and servicing a powerful but tricky electric motor… She could describe the whole process of composing a novel, from the general directive issued by the Planning Committee down to the final touching-up by the Rewrite Squad.
Hmm, even Orwell writing this book back in 1949 had the prescience to see where we are all headed. I wonder if he would be surprised at how quickly his fears have materialised. Marks looks at different sorts of writing, including screenplays, scholarly journal articles, and even gunslinger stories. He finishes his discussion of that third sort of writing as follows:
Back in 1960, when language and cognitive learning expert Jerome S. Bruner of Harvard was asked to comment on the MIT cowboy scripts, he said, “I have little doubt that we will be able to produce machines and computer programs that will behave in a fashion that we speak of [emphasis mine] as intelligent and that these will be of great aid to man…. Where my doubt comes in is whether we will be able to produce machines and machine programs capable of creative thinking.” Bruner knew whereof he spoke.
In this chapter the capacity of AI to write has been explored. Yes, AI can write – but it can write nothing deeply creative or belletristic. (p. 75)
Yes AI has done many things – some of them better than humans can do. But AI and machines will never be able to take the place of human beings – certainly not in the area of the arts and genuine creativity.
[1633 words]
