Students Who Got in Trouble for Using AI Writing

It’s ironic, many students have faced harsh consequences for using artificial intelligence in their schoolwork, yet at the highest echelons of our educational institutions, plagiarism seems to run rampant among top tier faculty.

If you've looked into the shifting ethics of AI and academic integrity, it only makes sense academia would have trouble embracing a new technology like generative AI. However, students that face harsh consequences for the use of AI today aren’t going to be comforted by the full scale acceptance of the technology in the near future.

I’ve scoured the internet to find tales of students using AI-generated text and getting accused of academic misconduct because of it. These are their stories.

Table of Contents

• Students That Got Punished

• Students That Actually Wrote Their Essays

• How Common Is AI Cheating in Schools?

• Do AI Detectors Work?

• How to Bypass AI Detectors

• Conclusion

• FAQ

Students That Got Punished

The most reported of these sad instances is the story of Marley Stevens, a college student who didn’t even use an AI chatbot to write her essay, she used Grammarly, an AI paraphraser, to proofread it. If you’ve read out blog post about AI paraphrasers, not only do you know they’re a flawed AI tool, but you’d know that being accused of academic dishonesty for using one is ridiculous.

Speaking of dishonesty though, if you delve deeper into this story, you learn that not only were her professors advocating for using Grammarly, but the university of North Georgia advocated for it on their own website, later taking down any mention of the software after Stevens’ accusation.

Marley was put in on academic probation that she eventually overcame, but she is far from the only student that encountered academia’s hypocrisy and incompetency around AI tools.

Another case involved multiple students at Yeshiva University that used generative AI like ChatGPT to write their take home finals. The students were caught when their English professor ran their finals through AI detection software and discovered the cheating. In response, the university implemented a whole new AI policy prohibiting the use of AI writing.

Reddit is full of these stories, here’s one below. In it, the student was threatened with possible expulsion but after pleading with faculty, her grade was taken from a zero to a C so she could pass the class.

Even more commonplace than punishment for students using AI writing, is AI detection systems screwing up and delivering false positives that sour college experiences.

False Positives and Bureaucratic Nightmares

UC Davis student Louis Stevens wrote a paper summarizing a court case for a class and hours after uploading it, not only did Turnitin flag a portion of the essay, but it was immediately forwarded to the Office of Student Support and Judicial Affairs.

After being subjected to an academic integrity review, Stevens learned that the university had just integrated Turnitin which was then spouting a 98% accuracy rate. Even though she proved her innocence, this wasn’t the first case of false positives at UC Davis.

Previously that year, William Quarterman uploaded his history exam to his professor who ran it through GPTZero and after it gave him a false positive, Quarterman was plunged into a bureaucratic nightmare before clearing his name.

These instances are certainly disheartening, but there was one case of false positive AI detection that was so egregious, it was actually funny. This was the case of a Texas A&M professor that temporarily failed his entire class after he ran their papers through ChatGPT and it told him that the papers were AI generated.

The class was quickly redeemed and the story goes on to be retold as an example of academic institutional incompetence.

Do AI Detectors Work?

There wouldn’t be a whole section about false positives if they did what they were supposed to do. Higher education simply doesn’t see another option. They don’t want students to skirt the learning process and not develop the critical thinking faculties that come with it. So, they employ bad technology. Even MIT published this report declaring that AI detectors don’t work.

As far as Turnitin goes, they claim to have a 4% false positive rate at the sentence level. That means for the millions of AI generated essays that get processed, there are countless false positives that are reported and just as many probably being neglected by faculty that understand the technology sucks.

The problem is simply in the algorithm. AI detectors are taught to check for the watermarks of AI writing, meaning simple and precise word choice, sentence structure, and sentence length, and it’s not like those can’t be exhibited in human writing.

Stanford even did a study that found AI detectors show a bias toward Non-native English speakers and international students for this reason.

As you can see above, each detector tested showed bias against non-native English speakers.

How to Bypass AI Detectors

In order to bypass AI Detectors like Turnitin, you need to use an undetectable AI service like StealthGPT. By humanizing AI text and removing the watermarks of AI writing, StealthGPT makes any content it generates human feeling enough for no person or detector to consider it AI generated.

If these students used StealthGPT, they would’ve never been caught in the first place, even when they wrote their own essays, StealthGPT knows how to avoid the pitfalls of AI detection that not even human writers are aware of.

If you want to learn how to bypass AI detectors with StealthGPT read our blog.

What to do if you get caught using AI

Lets say you’ve been accused of academic dishonesty. What do you do next? Here are the necessary steps to take to protect yourself:

• When responding to a failing grade or AI-usage accusation, be polite and assertive of your innocence.

• If this does not work, go to your student affairs office and get further direction.

• You will need evidence of your innocence so be prepared with documents showing your history with the assignment and preparation for it.

• Do your own research and compile sources that testify to the faultiness of AI detectors.

• Try to get witness testimony of your working on the project.

And remember, when you’re in front of a judicial committee, it’s often the case you can have a lawyer present if you feel that's neccesary.

Conclusion

When OpenAI created ChatGPT, they had to know students would use it for school work. Academia hasn’t exactly reacted to the technological shift gracefully, instead they’ve imposed restrictive measures and turned campuses into climates of suspicion.

Universities need to accept the fact AI will fully integrate into learning and teaching. Teachers will use it for everything from creating syllabi to curricula and students will use it for homework. The only way to ensure they don’t use AI is to host more in-person assignments, which you should fully expect to happen.

If you want to use AI writing tools for your essays, don’t be swayed by outdated notions of academic integrity to avoid using what technology is available to you. Work smarter and get StealthGPT to avoid AI detection in the first place.

StealthGPT bypasses every major AI detection software with unparalleled ease. We now offer a student discount that lets you subscribe to one of our plans at only .99 cents for the first month. Students that join StealthGPT become part of a community with thousands of users that have unlocked their potential using AI.

Don’t get left behind. AI doesn’t hinder your development, it lets you take proven shortcuts to success instead of wasting time.

FAQ

How Common Is AI Cheating in Schools?

There are multiple reports with varying answers, all of which are damning to academia for the frequency in which students use AI to generate their schoolwork. Here’s just one of many reports, this one claims only 20% of students say they’ve never used AI. Also, take into consideration that someone people wouldn’t tell you if they were polled.

With such a high rate of students using AI, it’s hardly fair to call them cheaters. They’re just young people growing up in a technological revolution is all.

Is there a free AI Detection tool?

Yes. Multiple free AI detection tools exist. The best of which is AI Detect because it also features a humanizer, making it a one stop shop for all students and content creators.

Is there a free Undetectable AI tool?

Yes. StealthGPT offers a free trial version to their software which allows you to use their AI humanizer and Stealth Writer for free and without any word limits. StealthGPT is not like other chat bots, we’re the kings of undetectable AI.