Admissions officer reveals how to spot ChatGPT in college essays in seconds

In today's digital age, the line between Google and ChatGPT has become increasingly blurred, as the AI chatbot seamlessly integrates into our daily lives. As more people turn to ChatGPT, a new trend has emerged—essays and content often carry a distinctive "ChatGPT signature," making it easy to spot when the AI has been used. Recently, a College Admission Officer from the University of California shared an insightful list of telltale signs that reveal when ChatGPT has written an essay.

Check out the entire Reddit post here:

In the past three days, I've reviewed over 100 essays from the 2024-2025 college admissions cycle. Here's how I could tell which ones were written by ChatGPT
byu/AppHelper inApplyingToCollege

The college admission officer notes, "Students from the United States, Europe, East Asia, South Asia, the Middle East, and South America shared their essays with me. But even among this diverse cohort, I noticed some striking similarities in their essays.”

"ChatGPT has a limited conception of what makes for a good essay, and with an uncreative prompt, it tends to make a 'safe' choice, which is often clichéd."?

As he points out, the context matters—your essay doesn’t exist in a vacuum and needs to stand out among countless others.

One of the clearest signs is the vocabulary used. "Certain words like 'delve' and 'tapestry' are far more common in ChatGPT-written essays," he observes. This becomes especially apparent when ChatGPT is used by students from certain backgrounds, like India, where these words would be uncommon.

Another giveaway is the overuse of extended metaphors. "ChatGPT's metaphors are usually not very original," the officer says.?

Common examples include comparisons to weaving, cooking, or painting—metaphors that are recognisable and often repetitive.?

"It’s mind-blowing that ChatGPT can understand and generate sensical metaphors, but they lack freshness."

Punctuation is another clue. "ChatGPT defaults to em dashes—like this—which are not widely taught in high schools," he explains.?

This, along with its use of straight quotation marks and curly apostrophes, creates a distinctive pattern that’s hard to miss.

Perhaps most revealing is ChatGPT's penchant for using rhetorical devices like tricolons. A tricolon is a rhetorical device involving three parts, for example - The Good, the Bad, and the Ugly or veni, vidi, vici.?

Credit: Unsplash

"If I see one tricolon in an essay, I'm not usually suspicious. If I see four or five, I can be almost certain ChatGPT had a 'hand' in it," says the officer.?

Clichéd endings are also a common ChatGPT trait. "ChatGPT writes very formulaic and clichéd endings," the officer remarks.?

It's a classic AI move: restate lessons learned, often with phrases like "I will carry this lesson into my future" or "As I grow, I will apply this knowledge."

The most telling sign, however, might be what the officer calls "Lord of the Rings syndrome"—the tendency to write multiple endings.?

"ChatGPT tends to give you several possible conclusions, which often contain the same tired phrases.?I could tell right away, and I’d bet most admissions officers could as well."

They note, "The best essays aren't mirrors—they're windows, revealing something unexpected, personal, and alive."

If the challenge is simply about crafting better prompts, one commenter shared a seemingly perfect solution on X: "Now, feed the above instructions to ChatGPT and tell it not to follow them."

What's your opinion on this?

To stay updated on the stories that are going viral, follow Indiatimes Trending.