(GoRealNewsNow.com) – Even though Artificial Intelligence (AI) has been of great assistance for some, new data has revealed that students sent in over 22 million papers likely written by AI in the last year.
According to a report from Wired, Turnitin, a service that checks for plagiarism, found millions of papers possibly packed with AI-generated content.
They launched an AI writing detection tool last year, training it on tons of papers from high school and college students and known AI-made content.
Now, the detector tool has a less than one percent false positive rate, even when checking whole documents.
Turnitin noted that its detector checked over 200 million student papers and found 11 % with at least 20 % AI-written content and 3 % mostly AI-written.
For instance, tools like OpenAI’s ChatGPT are big in academics, and students turn to them to cheat.
According to Breitbart News, a study found 17 % of Stanford students admitted to using ChatGPT in final exams. Elite Florida high school students were accused of cheating by using ChatGPT for essays.
Moreover, while educators want to hold students accountable for cheating, accurately detecting the use of AI in writing assignments can be difficult.
Notably, AI detection tools can pose a risk of false positives against English language learners. A report from a Stanford study found AI tools biased against non-native English speakers.
Turnitin claimed it trained its plagiarism detector on the writing styles of both English learners and native speakers. An October study listed Turnitin as one of the most accurate AI language detectors, Wired said.
Turnitin’s Chief Product Officer, Annie Chechitelli, suggested using the software to spark conversations about AI nuances, not just to pass or fail students.
“People don’t really know where that line should be,” she said.
Copyright 2024, GoRealNewsNow.com