By Annya Kong, Staff Writer

The AI digital image generator DALL·E’s take on the Great Gatsby. (Image created with DALL·E by Bridget Galibois)

As a society, we are constantly searching for ways to improve and innovate in the field of education. ChatGPT, the natural language processing (NLP) technology developed by OpenAI, has the potential to be a game-changer in this regard. 

One of the key benefits of ChatGPT is its ability to generate human-like responses to natural language inputs. This allows it to engage in real-time conversation with users, providing them with answers to their questions and engaging in meaningful dialogue. In the realm of education, this could be used to create virtual tutors or assistants that can provide students with personalized feedback and support.

However, it seems the currently free AI online chatbot is being used for more than just tutoring. By entering in a myriad of prompts, students are able to easily generate whole essays, which they can then use for convincing school assignment submissions. Don’t believe me? What if I told you those first two paragraphs were written entirely by ChatGPT?

When prompted to “write an essay about the themes in The Great Gatsby”, this was ChatGPT’s response:

“The Great Gatsby is a novel by F. Scott Fitzgerald that explores the decadence and excess of the Jazz Age. The novel is known for its exploration of themes such as the American Dream, the corrupting influence of wealth, and the destructive power of greed.

One of the central themes in The Great Gatsby is the American Dream. The novel portrays the idea of the American Dream as being corrupted by the pursuit of wealth and power. Jay Gatsby, the novel’s protagonist, is a self-made millionaire who becomes obsessed with winning back the love of his life, Daisy Buchanan. However, Gatsby’s pursuit of Daisy is ultimately fruitless, and he is unable to achieve true happiness or fulfillment. This portrayal of the American Dream as being hollow and empty suggests that the pursuit of wealth and power can ultimately lead to disillusionment and disappointment.

Another important theme in The Great Gatsby is the corrupting influence of wealth. The novel portrays the wealthy elite of East Egg as being corrupt and decadent, with characters like Tom and Daisy Buchanan and Jordan Baker being depicted as shallow and selfish. This portrayal suggests that the pursuit of wealth can corrupt and destroy individuals, leading them to become selfish and immoral.

A third important theme in The Great Gatsby is the destructive power of greed. Gatsby’s obsession with wealth and power ultimately leads to his downfall, as he becomes consumed by his desire to win back Daisy and loses sight of what truly matters in life. This portrayal of greed as being destructive suggests that the pursuit of wealth and power can ultimately lead to personal ruin.

Overall, The Great Gatsby is a novel that explores the decadence and excess of the Jazz Age and the corrupting influence of wealth and power. The novel’s exploration of themes such as the American Dream, the corrupting influence of wealth, and the destructive power of greed continues to resonate with readers today.”

The essay is not, by any means, inherently remarkable. It’s fairly simple, summarial, sourceless, quoteless, and not at all rigorous – but it was written in about thirty seconds, and took a thirtieth of the effort, and (with a little bit of editing) it passes. ChatGPT’s writing fits the typical high school or even college level classroom perfectly – and that, for teachers and administrators, is terrifying. 

Upon ChatGPT’s creation, an outpouring of articles of how “College Essays are Dead”; “The End of High School English”; “AI bot that can do schoolwork could ‘blow up’ US education system, with youngest at most risk: former teacher.”  In New York, public schools have banned ChatGPT on school computers, and similarly restrictive efforts have also been made in Seattle. Some hope to depend on anti-cheating tools like GPTZero, which is designed to catch AI-generated writing. Neither of these potential solutions, however, have the capacity to remain effective. Formermost, many students would still have access to the chatbot on their phones or their own computers. In this way, a ban on tech only hurts the underprivileged who do not have ready access to personal devices, widening the digital and educational divide. Meanwhile, as AI and AI-generated writing continue to rapidly improve, finding ways to catch or disrupt the use of these tools will also become increasingly difficult. In short, neither are a real, sustainable fix. 

So is that it? Is that really the end of the student essay? In some regards, yes. As what is currently an integral part of the standard U.S. curriculum comes under threat, it’s not hard to sympathize with the panic of educators as they rush to defend against AI. But while the immediate, fearful response is obvious, it would be worth the time to first ask if the scramble is even worth the effort. What is the purpose of the student essay – what makes it so worth preserving?

The invention of the essay is commonly credited to the Frenchman Michel de Montaigne. Essay in French means to attempt, and Montaigne used the term to describe an ‘attempt’ to put his thoughts into language. In other words, the fundamental purpose of the essay was to communicate ideas, not, necessarily, to create them. But now, ChatGPT can communicate better than most other humans. Though it may not be able to understand what it’s saying, the chatbot is able to concisely and coherently convey almost any message you ask of it, and the ability of artificial intelligence to do so can only improve over time.

This doesn’t mean the subject is obsolete. Saying that the invention of the calculator should have rid the math curriculum of teaching multiplication would make a pretty silly argument. Ultimately, even if machines can generate a better product, it’s often the key takeaways of a process which make a skill relevant. When asked about the purpose of the student essay, English teachers responded: “I want students to be confident in their ideas and their ability to express themselves,” said Ms. Margaret Howard. “I believe the purpose of the student essay is to formally assess writing skills developed, skills that need more development, and most importantly, to understand how students are thinking critically about texts,” said Ms. Tiffany Batiste-Gilmore. These are not unreasonable perspectives.

Even still, now that ChatGPT has automated much of what we think essential, the skill of knowing how to communicate in writing in and of itself has become far less useful. As a result, the standard of classroom essay writing – of generic prompts and specific formats – might, much like multiplication, be thought of as a basic skill – something to know how to do but, education-wise, be set aside to apply not as assignments in standalone entirety but as part of more complex works which necessitate such skills. In other words, ChatGPT could prompt a shift away from the heavy emphasis on essays in classroom writing and towards more individualized, varied, and in depth assignments.

As AI expands its scope and improves its accuracy, this will continue to apply more generally. The purpose of education is commonly touted as “to prepare students for the real world.” Well, the ‘real world’ is entering a new era. ChatGPT could destabilize white collar work. Some say the danger is exaggerated; for example, even though concerns about job displacement due to AI were prominent in radiology, there has only been an increase in radiologist job postings in the recent past, a one-shot negation of such fears. But artificial intelligence still had a transformative effect on the industry. As written in an editorial from the Radiological Society of North America, “Detection algorithms will solve “needle in a haystack” search problems […] And so AI will elevate the cognitive universe of radiologists to the top of their license — exercising judgment, creativity, and empathy as they interpret images in partnership with AI algorithms and patients. […] As we are lifted by the latest AI bubble, ‘Will AI replace radiologists?’ is the wrong question. The right answer is: Radiologists who use AI will replace radiologists who don’t. Similarly, artificial intelligence won’t ruin assessment, or render the current system completely worthless; like the essay, however, it might (happily) warrant a shift in focus. 

If school serves primarily as a stepping stone into a future occupation, then it’s important that education ensure that people grow to work meaningfully and optimally with modern technology, so that students can become the analogical ‘radiologists who use AI’. As such, it would be more prudent to implement ChatGPT into the classroom than to shun it. For example, students might be asked to critique AI writing, seeing its limitations and its benefits. Or, AI algorithms could be used to analyze submitted work and provide individualized feedback. Again going back to the essay, ChatGPT may allow students to dismiss the more rudimentary elements of writing, like structure or grammar, and focus instead on higher level and arguably more critical aspects like tone, voice, and the actual thought behind the attempt. One Payton student says, “When using resources like ChatGPT, I tend to use it as a way to establish and think out the structure of the essays I intend to write, since this is a thing that I usually get stuck on […] This also helps me learn what a typical example of a good essay structure is, which has allowed me to ease off of using this tool and develop my own structure.” Nate von der Sitt, a Payton senior, likens AI to typing or computer science: subjects that are taught because they’re important for the workplace. Because AI is “going to be a major part of the future […] schools should be teaching students how to use ChatGPT. Ignoring the skills required for the future because of prior beliefs is stupid.” 

If the fear is that this open availability of resources will result in a flood of AI-written essay submissions, then it might be the prompts –and priorities – that actually need revising. 

An anonymous Payton student said, “I’d use [ChatGPT] on assignments that don’t feel valuable to my learning or are very confusing/tedious.” This philosophy offers a different perspective of the issue. The straightforward perspective is that, of course, cheating is bad, and a fault of the student. Yet a more complex and more important problem arises if students don’t feel there is value in the work they are assigned; if they find the work repetitive or boring. In this case, even otherwise good students could feel compelled to save themselves some time. Even the ability for AI to generate a good (and thereby concerning) response is already indicative that the prompt itself is generic; it shows that there are similar essays that already exist, and enough of them to compile into something well founded. Technology like ChatGPT necessitates that instructors put more effort into crafting interesting, varied, and personalized assignments; this would not only improve the student’s perception of the work, but also make it harder to cheat regardless, as ChatGPT cannot draw from individual experience to write with originality. If AI-generated writing devalues work, then teachers could at least assign work with potential to be valuable. “I think it is perfectly fine for students to use ChatGPT,” one Payton freshman argued, “because we are just using resources that are available to us. Instead, teachers should assign assignments that can’t be AI generated.” 

To their credit, many Payton students insist that they would never use ChatGPT, citing it as “immoral,” “unethical,” and antithetical to “academic integrity” and Payton’s 4 C’s. However, even if Payton students are determined to abstain, eventually this will become harder to do than simply using AI.  “Though I’ve done my best to not use [AI] for now, there will be a time within the foreseeable future that it won’t be avoidable and remains the new standard,” admits Alexis, a Payton freshman. The developments before this point are critical, lest school curriculums and assessments really do become obsolete.

After ChatGPT’s release, to panic, to ban, and to discipline were all low-hanging fruit that gave easy answers to a complex problem. Articles and reports poured out detailing the new threat of AI and what it threatened to derail, but in many ways, the tremor could have just as easily been caused by fault lines that were already there. ChatGPT just happened to reactivate them. 

The issue is that the current school model mirrors, rather than complements, the foundational ideology behind machines. That is, that intelligence is rigidly defined, quantifiable, and based around knowing a lot of ‘right’ answers, and that learning can be standardized. As a result many of the prototypical in-school tasks that tend to center around algorithmic thinking, objective analysis, or blind procedure and repetition are becoming increasingly obsolete. It is these kinds of tasks which a report by the McKinsey Global Institute identified as at risk of becoming automated by 2030; at least one third of activities “in about 60 percent of occupations…implying substantial workplace transformations and changes for all workers.” On the other hand, specialization has become more valuable.

In this rapidly changing technology landscape, schools must help students take advantage of what humans still monopolize. In particular, the ability to explore, ask questions, and exercise creativity; to infer cause and effect relationships, and diversity of thought and individuality are all things AI struggles to, or fundamentally cannot, replicate. The general approach when it comes to education should therefore try to emphasize these qualities; giving students time to understand topics on a conceptual level, allowing them to explore their own interests within each subject, giving them the space to experiment with their own learning, and emphasizing creativity and intellectual risk-taking are some broad ways such a development might occur. More specifically: in high-school English, prompting students to express individual opinions or practice creative writing more frequently (Guess what? ChatGPT is pretty awful when it comes to generating poetry or good prose) as it relates to a given unit. Or, for math, a focus on concept-based rather than formula-based learning, such as a (guided) self-discovery of the ideas behind that mysterious little chant students know as the quadratic formula. This is not to say that standards or objectivity should be eliminated from the classroom, or that genericity inherently equates to worthlessness, but rather that educators should be critical about what is genuinely valuable to teach and what, whether it be student-driven or lecture-driven instruction, should be emphasized in different classroom contexts in order to promote meaningful learning.

In all honesty, the subject of AI is, in part, a guise to nudge at these suggestions. As a student myself, I’ve long wished that schools provided more space to express my own ideas, ask my own questions, and formulate my own conjectures. I wanted to learn naturally. I wanted to learn humanly. In the midst of academic rigor, I wanted time to actually think – about my own specific questions, about things that weren’t already written out. 

Ironically, commonly accepted Western notions of intelligence meet these desires. Ask people how they define intelligence, and the answers will range not only from knowing facts and being able to apply them – and, to clarify, by no means am I arguing that facts and procedures should not be taught – but also about imagination, about cleverness, about thoughtful deviations from the norm. They will say that specialization is more valuable than being well-rounded: when asked whether they’d rather know one subject in depth or know a lot of subjects at a surface level, people will often choose the former. Students are recurrently reminded that colleges look for a spike, and for something that sets an applicant apart. Critical thinking, intellectual risk-taking, “be unique”, “don’t be afraid to make mistakes,” “think outside the box”: all touted by educators with confident enthusiasm. In the sense of how we consciously perceive learning and intelligence, AI is far from a revolutionary catalyst.

Yet, despite our seeming agreement that qualitative creativity and quantifiable objectivity both occupy a valuable place in the intellect, the Western world has been seeing a shift towards analytical skills only, and the way secondary school operates reflects this ideology in a dangerously shallow way. Objective truth outweighs individual thought (students, when was the last time you were allowed to write an essay that used the word “I”?). The sciences are valued over the humanities. ‘Advanced’ classes, even when they are meant to attract students with specific interests in the study of the course, are more about covering more content quickly than covering less content deeply. Everything is a format, a rubric, a rule, a mnemonic, a number; there is no flexibility for thought, there is no need for an opinion, there is no incentive for risk, because at the end of the day everything comes to a score over 100 and a letter out of five. It’s mechanical but it worked, in its own way, in a macroeconomic way, for a while. It worked less well as technology improved and robots led to “robust negative effects […] on employment and wages” in the 1990s and early 2000s: blind procedures had started to mechanize. Now AI has continued this downward trend into the negatives, to the point where such a system would do direct detriment to students’ futures. You can no longer pit humans pretending to be machines against actual machines. Humans will lose. 

There is meaning to the current way we teach. Objective analysis and inductive reasoning have purpose. General application of old ideas is not less important than individual imagination for new ones. But neither is it more than, even less so now that AI comfortably dominates many aspects of the former category. It’s certainly not all there is, not even the majority of what there is, when it comes to being a human learner. ChatGPT has revealed the cracks in the system by widening them; the good news is that, on the flip side, it can also be used to repair them. We can give students a more relevant, more human experience of education by examining the effects of and conscientiously implementing AI – but it has to happen soon.