by Gabriel Sebastian
Who be the Monster?
We are in the middle of a technological and cultural evolutionary leap. The advent and proliferation of artificial intelligence brings with it incredible if not frightening applications and implications. Set aside China’s hubris or folly in naming its AI surveillance network “Skynet” to consider more mundane and practical AI applications: editing, writing, image creation. (But maybe even problem solving? City planning? Therapy? Medical advice?)
Here at confetti, and in our adjacent writers’ group, conversations often turn to the use of language learning models, virtual assistants, A.I., and how our plucky little magazine, writers and poets, and society itself will deal with the looming, algorithmic giant.
How will confetti deal with A.I. submissions?
How would we know if submitters have utilized A.I. to write, to correct, to make the themes, or generate ideas of a poem or story? How would we know if they used A.I. to create their entire work?
How would we even know if an artificial intelligence itself submitted a poem or a story?
Recently, a few dozen similar-looking submissions gave us pause and forced us to have real conversations about how to deal with “the A.I. problem.” (It turned out the submissions were part of a class project, all bearing similar formats, subject matter, and origins.) Even considering a genuine misperception, what we surmised and projected at the precipice of this leap and the questions we asked felt Impending if not imperative.
Some have suggested using A.I. programs to detect if a text is A.I. generated. A virtual fox guarding the hen house against other virtual foxes, eh? However, one blogger ran the Preamble of the U.S. Constitution through an A.I. detection program, and it told him the document was 100% A.I. generated! (Benj Edwards wrote his thought-provoking, “Why AI writing detectors don’t work,” where he asks, “Can AI writing detectors be trusted?” Find article here.)
One of our editors also asked, “Can we trust A.I. to tell us if something is generated by A.I.?” And I retorted, “Can we trust A.I. to tell us the truth?” knowing damn well that even though humans write the laws and programs, we are still vulnerable to the (Isaac) Asimovian conundrum: What do we do when a machine outgrows its programming? At least, besides, there is the notion that flawed programmers beget flawed programs. Or an even more frightening aim: that of A.I. being able to program itself?
And what of copyright? The plagiarism of ideas? Doesn’t A.I. draw from already published sources, copyrighted or otherwise? (I still laugh thinking of A.I. as the infamous monkey on a typewriter typing for infinity and creating all of Shakespeare’s works.) Will simplicity of use eventually override ownership?
Or worse, does A.I. make it easier to regulate and restrict, feed propaganda, impose control, and ultimately kill free thinking and free speech?
We know high schools, colleges, and universities all have various practices and prohibitions on the use of A.I., rooted mostly in their plagiarism policies, yet we have heard that some dare to allow students free use of this device. We wonder at the effect of this new tool.
I asked an A.I. what it thought of the problem of A.I. submission:
(MY QUESTIONS) Why would an A.I. company or an A.I. bot want to submit poetry and art to an online magazine? Has this ever been done before?
(THE RESPONSE) There are several reasons why an A.I. company or A.I. bot might want to submit poetry and art to an online magazine:
Demonstrating creativity: By submitting poetry and art, an A.I. company or bot can demonstrate its creative capabilities and showcase its ability to generate original and meaningful content.
Exploring human-A.I. collaboration: Submitting poetry and art can be a way for A.I. companies or bots to explore the possibilities of human-A.I. collaboration and to see how their generated content is received by human audiences.
Testing and refining algorithms: Submitting poetry and art can provide a way for A.I. companies or bots to test and refine their algorithms, by seeing how their generated content is evaluated and critiqued by human editors and readers.
Raising awareness and sparking discussion: By submitting poetry and art, an A.I. company or bot can raise awareness about the capabilities and limitations of A.I.-generated content, and spark discussion about the role of A.I. in creative industries.
As for whether this has been done before, the answer is yes. There have been several instances of A.I.-generated poetry and art being submitted to online magazines and literary journals.
For example, in 2016, a team of researchers from the University of Cambridge submitted a collection of A.I.-generated poems to the literary journal, The Guardian. The poems were generated using a neural network algorithm and were designed to mimic the style of famous poets such as Shakespeare and Keats.
In 2018, an A.I.-generated novel titled, “The Day a Computer Writes a Novel” was submitted to a Japanese literary competition and passed the first round of judging. The novel was generated using a combination of natural language processing and machine learning algorithms.
More recently, in 2020, an A.I.-generated art piece titled “Edmond de Belamy” was sold at a Christie’s auction for $432,500. The piece was generated using a generative adversarial network (GAN) algorithm and was designed to mimic the style of 19th-century European art.
These examples demonstrate that A.I.-generated poetry and art are being taken seriously by the literary and art worlds, and that there is a growing interest in exploring the possibilities of human-A.I. collaboration in creative industries.
Shall A.I. be the death of creativity?
I recently posted the above question on our Facebook group, though none were inclined to respond with anything other than an emoji. Yet, this is a pertinent question in the face of the speed and ease of A.I.
Should we be worried lest A.I. become an echo chamber spitting back familiar phrases and recycled ideas at us, gleaned and regurgitated from the ethereal Internet?
Perhaps the worst outcome of A.I. (other than our typical Terminator Skynet fear) is that it does become a dreaded echo chamber, feeding upon itself, cannibalization of the worst kind, and we are left swallowing our own tales, until there is nothing left but literary stagnation and creative paralysis.
Or maybe, hopefully, A.I. will free human creativity from the shackles of language, supporting those troubled by language, and allowing the human spirit to flourish, enhancing communication, pushing us to push past our linguistic limitations.
One contributing poet to confetti recently shared an amazing poem entitled, “Luddite 2024.” In a conversation with poet J. Clifford Milligan, he assured me that the invention of a kitchen knife when humanity was using rocks to pulverize food should not be viewed as “cheating” or voodoo, but as a time saving tool enhancing human culinary arts. Nor did the invention of the lighter bode the downfall of humanity as opposed to rubbing two sticks together. He introduced me to the term ‘Luddite,’ and in typical rule-of-three fashion, the concept seems to return to the zeitgeist, and I notice it everywhere lately.
Artificial intelligence could be the enhancement elevating literacy and human intelligence, pushing us to further and better means of communication, meaning making and expression.
There’s probably no real harm, right? Plagiarism or copyright infringement doesn’t seem likely and with A.I.’s speed and reach, certainly more difficult. Besides, in the song “Cemetery Gates,“ The Smiths’ Morrissey croons this spot-on warning: “Cause there’s always someone, somewhere, with a big nose, who knows . . .”
One conclusion reached—the editors and selectors at our magazine will rely on our intuitions, experiences, and wisdom along with the discernment of the derivative, and we will still seek stories and lyrics that speak to the human condition and exude merit. And if sentiment be filtered through the digital heart of artificial intelligence, and we are fooled or blinded ourselves overly sensitive, then our error be only too human.
The onus of copyright and originality rests with our submitters, as we ask only for “original” works, where copyright is owned by the submitter, and due credit is given where necessary. We will endeavor to ensure we have cited appropriately and will remove any contested content.
A.I. does not then frighten, nor subjugate (yet) as long as humanity still controls the keyboard, as long as we remain creative and hungry, as long as we harness the tool and let it not harness us.
Postscript: Except as noted above, no AI was utilized in the generation of this essay.
© 2025 Gabriel Sebastian All rights reserved.
Click or tap here to see Gabriel Sebastian’s profile.
Use the “Leave a Comment” form below to submit comments on this piece.