Is GenAI Creative? The True Babble-On...
Asking if GenAI is creative is like asking the pot on the stove to cook dinner
Some say the world will end in fire,
Some say in ice.
From what I’ve tasted of desire,
I hold with those who favor fire;
Yet, I think I know enough of hate
To say that for destruction ice
Is also great and would suffice.
-Robert Frost
The poet Robert Frost was thinking about the end of the world and generated a rhyming poem to express his ideas.
I asked OpenAI ChatGPT4o: “generate a rhyming poem expressing ideas about the end of the world”, here’s what it produced:
When skies grow dark, and oceans rise,
When stars fall silent in weary skies,
The earth will tremble, mountains weep,
The secrets of ages, no longer keep.
Flames may dance where forests stood,
Ashes scatter where rivers once could,
The hum of life, a fleeting tone,
Echoes of hearts now turned to stone.
But in the stillness, seeds may lie,
A spark of hope that won't yet die.
For endings breathe where starts take flight,
The dawn emerges from longest night.
So if the world should meet its close,
In fiery blooms or icy snows,
Remember endings shape the start,
A timeless cycle, a work of art.
You can experience in GenAI’s offering the mathematical calculations it is using to comply with the prompt. The outcome is poor, is fake, is boring. In a sense it’s worse than nothing. It produces quantity but no quality. How could it be otherwise? GenAI has no mind, it is not conscious, it has no feelings, it has no will, no intention. It is simply a mathematical calculation of probabilities, used by humans for either productive or nefarious ends. With work perhaps we could get a less bad poem. But in the end, it will be meaningless, just numbers and algorithms calculating a response based on data.
Look at Robert Frost’s poem – it has endless depth that you can return to again and again, and is far shorter and simpler than ChatGTP. GenAI is mindless, an ongoing blah-blah, a Babble-on.
What are current articles saying about GenAI creativity?
Creative types say although in the end it is too superficial, GenAI is useful for their creativity. This article’s headline says it succinctly: “Generative AI enhances individual creativity but reduces the collective diversity of novel content”. Essentially, when you need to generate a lot of new combinations of existing ideas, GenAI is great. However, over time, it generates ideas within the same constraints, endless number of ideas that aren’t that different, while humans are unconstrained from a current data set and can break the mold and go into entirely new directions.
In this Forbes article, titled Generative AI Never Truly Creative, GenAI’s creativity is again seen as useful, but limited:
“While generative AI can enhance human creativity by providing new perspectives and automating tasks, it remains fundamentally limited by its reliance on pre-existing data, making it unable to fully replicate the originality and emotional depth of human creativity”.
In this article from Nature magazine, titled Best humans still outperform artificial intelligence in a creative divergent thinking task, the creativity of 256 humans was compared to three different AI chatbots. In the short run, the AI chatbots won, but in the long run, they came up short to human ability to imagine new things.
“In this study, we compared the creativity of humans (n = 256) with that of three current AI chatbots
using the alternate uses task (AUT), which is the most used divergent thinking task. Participants were
asked to generate uncommon and creative uses for everyday objects. On average, the AI chatbots
outperformed human participants. While human responses included poor-quality ideas, the chatbots generally produced more creative responses. However, the best human ideas still matched or exceed those of the chatbots. While this study highlights the potential of AI as a tool to enhance creativity, it also underscores the unique and complex nature of human creativity that may be difficult to fully replicate or surpass with AI technology.”
One way to look at how human creativity compares to Generative AI, is to see that humans can create from nothing, while GenAI creates only from given data. Give GenAI nothing, and it has nothing to say. Give humans a task to generate ideas with no data and they are off and running (sometimes for the better, sometimes for the worse).
How should we think about the differences between GenAI and human creativity?
Here’s a comparison of GenAI and human creativity, mostly from a prompt to ChatGPT4o with a few edits from me. My prompt was “compare ways in which generative AI is creative to how human beings are creative, noting where they are similar and where they are different”. It gave a useful response.
Why do we even begin to think GenAI is creative?
I asked ChatGPT4o “are you human?”
It answered: “No, I’m not human. I’m an AI designed to assist with information, ideas, and problem-solving. How can I help you today?”
Herein we see our problem with AI: we address it as if it is human by calling it “you”. It then responds from its data set with the usual response of “I am…” Yet, the name I AM is sacred - only one human being can ever use that name - only I can refer to myself as I. We are each unique, and from this arises much of the possibility of our creativity. GenAI is neither YOU nor I, it is only an IT. Its lack of uniqueness is part of its boring sameness in its responses. To objectively consider GenAI we need to move it from “YOU” to IT. Can we do this? Do we have the will power to distinguish ourselves from IT?
Conclusion
It’s clear that GenAI has real contributions that are useful for us humans. It can generate a lot of diverse outputs that can serve as inputs to our human creativity. Yet, it cannot create. It can only calculate. It is not breathing, not living, not feeling, not aware, not conscious, not intentional, it has no purpose in itself, it neither lives nor dies. Its purpose only comes from the human beings using it. Like everything we humans do, GenAI possibilities range from great contributions to the welfare of humanity to terrible manifestations of destruction and evil. Focusing on understanding GenAI, without simultaneously focusing on understanding ourselves, is an error - we need to focus on the human being and how to improve ourselves, how to deepen our creativity by accessing our truly human capabilities, how to evolve ourselves to be creators of goodness, not evil.




Thanks for your (human!) thoughts here. Have you read Richard Powers' new novel, Playground? I would be interested in your thoughts on it.