With the increase in online learning, so too has increased the potential for cheating and plagiarism in today’s classrooms. As a result, educators have been forced to find innovative ways to protect their assessments from any malpractices by students. ChatGPT might be one of those solutions, promising teachers an easier way to detect cheaters among their pupils.
OpenAI recently startled the learning world with its open launch of ChatGPT, a powerful AI chatbot that can remarkably craft texts for customers like students and workers. Just two months after its initial release, the company is now giving educators another treat: a useful instrument to help them accommodate this technology into their workspace.
OpenAI unveiled an AI text classifier that detects whether a human or AI wrote a text. Although admitting the tech is still far from perfect, they rolled out this groundbreaking feature to provide an initial step into something more comprehensive in the future.
Using the new tool, it is possible to take an input, such as a school essay, and assign it to several categories using a machine learning system. This system works with English AI-generated text, giving an outcome ranging from “very unlikely” to “likely generated by AI.”
Educators have requested that such a ChatGPT feature can help, according to the policy research director of OpenAI, Lama Ahmad, who said this to CNN. Despite its benefits, he also added a warning and stated that it should be used cautiously.
Lama Ahmad says:
“We really don’t recommend taking this tool in isolation because we know that it can be wrong and will be wrong at times – much like using AI for any kind of assessment purposes.”
“We are emphasizing how important it is to keep a human in the loop … and that it’s just one data point among many others.”
Teachers have commonly utilized past student work as a gauge to assess if the writing was comprised of an individual. This new technology supplies another benchmark for the conclusion formulating about the text’s authorship.
Ahmad went on to say:
“Teachers need to be really careful in how they include it in academic dishonesty decisions.”
ChatGPT continues to prove its worth; it has created unique narratives, composition pieces, and tunes and passed various tests, including some held at Minnesota and Wharton. Notably, scientists have also been surprised by the preciseness of laboratory work outputted from the platform.
ChatGPT has recently gathered widespread attention in the educational community, but not without raising some alarms. Its usage has been banned by public schools in New York City and Seattle on networks and devices due to concerns that it may disrupt student learning.
Educators are rapidly changing their assignments due to ChatGPT, though the degree to which students have adopted it and its potential detriment to education are yet unknown.
OpenAI has joined the group of initiatives that set out to recognize written pieces dispersed by ChatGPT. For instance, Turnitin is focusing on built detectors to detect plagiarism from works done by ChatGPT, helping educators realize when student assignments are made using the tool.
Edward Tuan, a Princeton student, revealed to CNN that the Beta version of his system—ZeroGPT—is met with tremendous interest by teachers as virtually 95,000 people have already tested it. He further elucidates that the system is a ChatGPT detection feature.
On texts shorter than 1,000 characters, OpenAI admits its AI content classifier displays a “very unreliable” detection prowess. Moreover, in these circumstances where answers are predetermined, for instance, a list of the initial 1,000 prime numbers have indistinguishable sources, whether machine-generated or human-crafted.
Despite its imperfections, ChatGPT is a tool that could help catch cheating students. In a world where online education is becoming increasingly popular, tools like this will become increasingly necessary. Cheating is an issue that has plagued education for centuries, and it will be interesting to see how ChatGPT develops in the future.