Can artificial intelligence capture something written by artificial intelligence?
Haebom
2
3
1
Created by
Haebom
Created at
Marley Stevens (21), a junior at the University of North Georgia majoring in Human Services Delivery and Administration (BS), used the web browser extension Grammarly to proofread her assignments.
Grammarly is a service that corrects English spelling, grammar, etc., and recently introduced an AI-based editing function. It is a service that I have personally been using for several years. (Highly recommended)
And then, a few days later, I received an email from the professor in charge, and the email was a one-sided notice that the assignment Stevens had submitted was determined to have been generated by AI, and therefore, the assignment would be given a zero and the course would be failed. In short, they were disciplinary actions for cheating.
Stevens’ position is very unfair. She claims that she was a student who had maintained a 3.0 GPA in the past, that she spent a lot of time on this assignment, and that she only used Grammarly to correct punctuation and spelling. Of course, the professor ignored this claim.
Stevens' Tiktok is complaining of injustice by revealing the email he received from the professor...
Here, Stevens took to TikTok to share his plight, filmed a video demanding the retraction of the unfair punishment, and reached out to Grammarly HQ for help. Grammarly was quick to respond, stating that “suggestions for spelling, grammar, clarity, conciseness, and tone are not driven by generative AI,” adding that some tools that detect AI-generated content are not 100% accurate.
The professor couldn't just sit still, so he emphasized again that he had used the plagiarism prevention software TurnItIn.com and that he had no intention of retracting the already imposed disciplinary action. Well, for now, Stevens was removed from the scholarship list and the disciplinary action process began.
In fact, various domestic and international AI-generated content detection services are being released. Those who know the principle will know that this is quite inaccurate. It is possible to mark the content as being created by AI with probability, but if they insist that it was created directly by humans, there are not many ways to find out. This is even more so unless the entire process of creating this content is monitored or watermarked. This incident is worth paying attention to as it is similar to TurnItIn VS Grammarly as each tool is disclosed. (It's like a battle between a spear and a shield...)
It is a battle of truth, and professors and students will engage in a fierce debate, but this incident is also the most superficial manifestation of an important discussion about how far AI-generated creations should be accepted academically and how to find out about them.
Recently, there have been various cases of abusing in academia, and there are many cases where this actually happens, so it seems interesting to see how to catch and run away from this. Personally, I think that in the process of catching and running away like this, technology and methods also develop... And as I read, there is a generative speech style?, but I think that even this will actually disappear in a few years, so it seems a bit difficult to catch it by speech style or pattern, and there must be a way to catch it through the writing time or process, but this is a difficult problem because it involves personal privacy and security.
Subscribe to 'haebom'
📚 Welcome to Haebom's archives. --- I post articles related to IT 💻, economy 💰, and humanities 🎭. If you are curious about my thoughts, perspectives or interests, please subscribe. Would you like to be notified when new articles are posted? 🔔 Yes, that means subscribe. haebom@kakao.com
