Will AI Replace My Creative Writing Degree?

So there you were last fall, comfortably ensconced in your creative writing degree program, eyeing a career — or at least a side hustle — as an author, screenwriter, copywriter, or any other endeavor rooted in the written word.

Then ChatGPT happened. The November 2022 launch of the artificial intelligence (AI) chatbot turned the creative world upside down.

Will this new tool constitute a godsend, a fount of inspiration and ideation destined to make writing tasks easier? Or is it the Grim Reaper come to collect humanity’s creative soul and terminate any aspirations you may harbor for a writing career?

In other words, will AI render your degree obsolete?

Given that the most celebrated manifestation of AI, ChatGPT, produces written content on demand, it’s no surprise writers rank high among the professionals most affected by this technology.

A recent study by OpenAI (which owns ChatGPT), OpenResearch, and the University of Pennsylvania found that creative writers and authors have extraordinarily high “exposure” to disruption from AI tools.

This new reality caught the attention of the Writers Guild of America, whose members went on strike in May. Among other concerns, the union aims to “regulate use of material produced using artificial intelligence or similar technologies.”

Could AI replace Hollywood writers? To test this theory, I asked ChatGPT to write me a “Ted Lasso” script. Now that the series has ended, perhaps fans can binge on bot-written episodes.

Here’s what it offered:

And so forth. It went on for some time, though I didn’t stick around to see the rest.

Presumably, clicking “regenerate” would result in yet another script. Do that 11 more times and you’ll have a full season, though one no doubt comprising a series of unrelated stories and not a continuous arc.

Still, it’s easy to see what’s causing all this consternation. Assuming a showrunner could feed the bot specific details and curate the results until a feasible plot unfolds, complete with realistic dialogue, the role of professional writers would significantly diminish.

At the very least, chatbots could rewrite certain scenes expeditiously and, with some guidance, expertly.

“That’s the nightmare scenario,” John August, who’s on the Writers Guild negotiating committee, told The New York Times.

Michael Schur, creator of “The Good Place” and co-creator of “Parks and Recreation,” harbors a similar fear of AI interference.

“It is not out of the realm of possibility that before 2026, which is the next time we will negotiate with these companies, they might just go, ‘You know what, we’re good,'” he told the Times. “We don’t need you. We have a bunch of AIs that are creating a bunch of entertainment that people are kind of OK with.”

Although that’s the rub. Will AI material be just “good enough” to pass muster with an audience? Or will it seem derivative, largely because AI doesn’t create anything new but rather repackages and regurgitates what’s already been produced?

The answer, it seems, involves the notion of what, exactly, it means to be creative.

No doubt AI chatbots can create content, but are they creative in the ways humans are, perhaps even moreso?

In a New York Times op-ed, famed linguist Noam Chomsky and his colleagues note the advent of tools such as ChatGPT have signaled the arrival of anthropomorphic bots capable of human thought processes.

Some warn we’ve reached “that long-prophesied moment when mechanical minds surpass human brains not only quantitatively in terms of processing speed and memory size but also qualitatively in terms of intellectual insight, artistic creativity and every other distinctively human faculty,” they wrote.

Initially, this seems preposterous. Machines can’t think; they only process. Certainly, no bot could, for example, produce poetry capable of fooling literary experts, could it?

Well, it turns out one could. A team from Microsoft and Kyoto University created an AI poetry tool and tested the results using Amazon Mechanical Turk, a crowdsourcing website ironically claiming “there are many things human beings do much more effectively than computers.”

The experts, the study found, were “fooled more often than not.”

That’s because we can, in fact, “teach” a computer to be creative, claims Craig Wisneski, co-founder of the AI platform Akkio.

“Given the tremendous rate of progress we’re seeing in AI,” he wrote, “we can naturally expect AI’s creativity to continuously improve to the point where AI-created art could be indistinguishable from human art in any area.”

However, Wisneski added, computers aren’t able to create truly random data. And randomness, producing something truly novel, lies at the heart of human creativity.

“Humans can create and perceive patterns from (apparent or real) randomness, because humans, for better or worse, see patterns everywhere,” Wisneski wrote. “But computers can only work in existing patterns; it’s actually impossible for a computer to internally create truly random — and therefore truly original — data.”

Take ChatGPT. Its algorithm sorts through millions of sources searching for the next most likely word based on patterns it has “learned,” creating strings of text that seem logical and plausible.

Yet true creativity is not using that next most common word but instead finding new pairings and sequences, producing that “spark of genius” that connects ideas with surprising, not predictable, results.

“The key characteristic of AI’s creative processes is that the current computational creativity is systematic, not impulsive, as its human counterpart can often be,” Chloe Preece wrote in The Conversation. “It is programmed to process information in a certain way to achieve particular results predictably, albeit in often unexpected ways.”

Any AI attempt to create will be “constrained by the bounds of its training data,” notes the futurist Thomas Frey, “making it less likely to generate truly original or groundbreaking ideas that have not been seen before.”

A counterargument might suggest all creativity is simply iterative and derivative, building on existing human knowledge and understanding.

We all have frames of reference we draw upon when creating something “new.” Writers use the same 26 letters, at least in the English language, and operate within an established vocabulary codified by the dictionary. (Devotees of constructed languages might beg to differ, however.)

Consider the truly revolutionary inventions in history: the printing press, the light bulb, the telephone, the automobile, the steam engine, the airplane, the computer, the internet, the pet rock. None of these sprang from whole cloth but rather manifested when people combined existing materials to incrementally advance the boundaries of knowledge.

How do I know this? ChatGPT told me.

I asked the bot if it planned to replace my creative writing degree (for argument’s sake, let’s say I have one). Its reply:

Even the almighty bot concedes.

Instead of fearing AI, think of it as a companion tool augmenting your work, suggests Wayne Butterfield, a partner at ISG Automation.

“Rather than put people out of work, using AI alongside their own skills enables even jingle writers, artists, etc., to do more, faster and potentially cheaper,” he said.

Because ultimately, AI isn’t sentient. It can’t feel or emote. It can’t express what it means to love, to yearn, to mourn, to exalt. It can borrow another’s words capturing those experiences, but that expression will ring hollow without a human author.

In a world filled with AI content, consumers will gravitate toward the authentic and pay a premium for it.

So rest easy and continue pursuing your bachelor’s or master’s degree in creative writing or your MFA in fiction or poetry or nonfiction. In the battle against bots threatening to reduce creativity to algorithmic formulas, we need your help.