The Risks of Utilizing AI for Writing Essays

As a professional writer, I could talk endlessly about why you should avoid letting machines take over your writing tasks, but that's not the focus of my discussion today.

I genuinely enjoy writing; for me, it's an enriching experience that brings joy, pride, and serves as a platform for self-expression. But I understand that writing doesn’t come equally easy for everyone. Students, in particular, often find the seemingly endless stream of essays for English classes and general studies to be quite daunting. I can relate to that feeling.

Yet, even for those who may not relish the process, it's important to recognize the value that writing offers. Those essay assignments aren't just there to complicate your life; they're designed to cultivate skills that employers are desperately seeking today: analytical thinking, problem-solving abilities, creativity, and the capacity to approach challenges from diverse perspectives. Writing is essentially a form of thought processing, and the more you engage with it, the more adept you become at forming and articulating your ideas.

But you're likely already aware of this, even if you don’t fully acknowledge it. Perhaps you find yourself relying on AI to produce essays you deem unimportant, or you resort to it when you're racing against the clock to submit an assignment and maintain your GPA. We all face challenging circumstances.

That’s why I opted to take a practical approach and explore the pitfalls of using AI for crafting essays in an academic setting. The flaws inherent in these AI tools can really get you into trouble, and in critical moments, it's often wiser to hire someone to handle your assignments than to depend solely on AI. paper writing service And let me clarify, not all writing services will drain your wallet. There are many platforms available that can either write essays free of charge or provide you with tools to expedite your writing process and refine your drafts seamlessly. You have a plethora of better choices compared to AI, and I’ll explain why.

AI is unreliable

This subject has seen its fair share of discussions among educators and journalists alike, but it bears repeating due to its significance. ChatGPT and similar generative AIs are systems crafted to produce text based on existing data. Their expertise lies solely in text manipulation - they don’t comprehend ‘reality’ as we do, and their grasp of facts is limited to what's articulated in the training materials. AI lacks the ability to verify its sources or distinguish between accurate information and misinformation. Its function is primarily to summarize and create coherent narratives in response to your inquiries. If you expose AI to textbooks and encyclopedias, it may serve as a useful educational tool. But if it's trained on the sprawling content of the internet, the result can be intriguing yet often misguided.

In several tests, I had AI attempt to write essays on sophisticated college subjects and instruct it to provide citations for its claims. There were moments it didn’t follow through, while at other times, it managed to list sources that appeared academically credible. The catch is that while the names of the experts and journals were genuine, the articles it cited never actually existed – the AI fabricated them based on patterns it learned.

Thus, the first takeaway is that AI isn’t actually capable of composing essays. It can mimic the structure of an average essay but ends up inventing facts and blurring details. Validating these claims is akin to detective work that requires ample time, which is precisely what you're trying to save by using AI in the first place.

AI is detectable

This remains a matter of contention, but tools designed to detect AI-generated content usually can identify when a text has been produced by an algorithm. These detectors operate on probability, hence they cannot provide absolute certainty. Their use in educational institutions stirs some debate. Numerous reports have characterized cases where students faced accusations due to their essays being flagged as AI-generated, even though those essays were perfectly average.

The crux of the issue often lies not in the AI detectors themselves but in the shortcomings of academic officials who fail to carefully consider each case, opting to jump to conclusions without thorough investigations. In some instances, institutions have even neglected to seek additional assessments from other detection software before taking punitive actions against students. This is a pertinent point and could serve as part of your defense if you find yourself questioned about the originality of your work.

Nevertheless, if a number of different detectors flag your work as AI-generated, that does raise eyebrows and warrant further scrutiny. If you did utilize AI for your essay, it’s highly likely you will struggle with additional evaluations, like being asked to elaborate on your topic or to submit drafts, notes, or other pieces of genuine research. If your sole argument rests on the fallibility of AI detectors, you're likely setting yourself up for failure. Even if immediate repercussions like suspension or expulsion don’t happen, expect to receive a warning, and find yourself kept under close observation by school authorities.

AI is repetitive

The AI’s inability to demonstrate a distinctive style and finesse in writing brings forth another issue altogether. Its tendency to deliver redundant passages that essentially reiterate the same points in varying phrases can be infuriating, and I assure you, your professor will likely feel the same frustration. As previously mentioned, AI can't distinguish between real-life concepts and the words describing them. If something is described using substantially different terminology, AI assumes it represents entirely distinct ideas. For example, it may frame 'feeling revitalized' and 'having increased stamina' as two separate benefits of substituting water for coffee or categorize 'communication', 'collaboration', and 'teamwork' as three different elements of group projects, even if the underlying examples are identical.

While this redundancy might not scream 'AI-generated text,' it certainly casts doubt on your comprehension of the subject matter. If you find that your persuasive essay reiterates the same argument in three different ways, that raises a flag about your level of understanding. Such writing is unlikely to earn high marks.

AI is not original

I want to clarify: I'm not critiquing AI for lacking creativity or for failing to contribute groundbreaking ideas—I’m addressing the issue of plagiarism. Identifiable segments of unoriginal text are at the heart of this concern.

One of the claims made by champions of AI-assisted academic dishonesty is that essays produced by ChatGPT are notably original. This suggests that the algorithm is so adept at rephrasing that the origins of its facts and assertions are impossible to trace.

However, this assertion only holds up to a point. Despite the impressive variability of ChatGPT’s output, it still relies on a limited set of structures and synonyms. Sure, the earliest AI-generated essays might have seen 0% plagiarism scores, but that's simply because they were unique productions. Don’t expect a hundred essays on a well-trodden topic, all crafted by AI using the same datasets, to show no similarities. It turns out, AI has its set of common phrases and expressions that often recirculate during plagiarism checks.

Are you skeptical? Pick any AI platform you prefer and ask it to produce a blog post on a random subject. Then run that output through any plagiarism detection software. There’s a significant chance that an article on the same topic published within the past year will turn up, revealing similar structures and suspiciously alike phrasing. Certainly, the plagiarism percentage might not be alarmingly high, but as more individuals begin to exploit AI lazily for generating generic blog posts filled with 'effortless original content', you can anticipate a surge in articles with increased overlap.

AI-generated writing will consistently land in the realm of mediocrity at best.

Even though AI-generated content might appear quite polished to novice writers or language learners who haven't yet mastered their target languages, it ultimately falls flat when assessed critically. By its very nature, AI outputs are average—they represent a calculated average based on the extensive data it has absorbed. Regardless of what promotional material claims, it's far from 'excellent.' In fact, calling it 'good' would be a stretch. Consider it merely 'acceptable' at its peak. Are you truly satisfied settling for such mediocre results? Don’t you aspire to enhance your skills and push your limits? Perhaps I’m framing this incorrectly—are you genuinely content with just scraping by with a C-?

The prospects for improvement in this area are minimal. AI's capabilities are inherently tied to the information we creative, adaptive humans provide. If an increasing number of people opt to forgo writing in favor of quick grades, clicks, or social media likes, the wellspring of AI’s creative potential will quickly dwindle. Fast forward three years, and we may find ourselves clicking 'Regenerate' in hopes of generating something anew and remotely original, something that hasn't been regurgitated multiple times already. Unfortunately, this endeavor will prove as challenging as extracting water from a stone.

Relying on artificial intelligence to craft academic papers is generally unwise. These generated essays often lack depth, tend to be redundant, and might contain inaccuracies or fictitious references. They're also easily detected by AI and plagiarism detection tools. However, the possibility of being caught shouldn't be the primary reason to avoid using AI for writing assignments. At its core, writing serves as a means of communication. We express ourselves through writing to share our perspectives, advocate for our beliefs, persuade others, and collaborate in shaping the world to reflect our ideals. If we merely exchange AI-generated text that no one finds engaging, we lose out on genuine interactions—there's no discourse, no disagreement, and certainly no growth. Isn't that a rather dull existence?