OpenAI Rolls Out New Feature to Help Teachers Crack Down on ChatGPT Cheating — But Admit the Tool Is 'Imperfect'
Amid increasing concern regarding the use of ChatGPT by students, the tool's creator has announced a new feature that could help teachers spot it in academic context.
By Madeline Garfinkle •
Since its launch in November, ChatGPT has disrupted various industries — from real estate to law. However, in the context of academia, teachers have growing concerns about whether using the tool is considered cheating.
ChatGPT, created by artificial intelligence company OpenAI, has the power to create essays, poetry, draft legal documents and more when given a prompt. While the results may need some editing, the tool's efficiency has garnered worldwide attention for its accuracy.
Public schools in Seattle and New York City have already banned the use of the tool over cheating concerns and its power to disrupt genuine learning. Now, OpenAI has announced a new feature that may help teachers spot the presence of ChatGPT in essays and other assignments, CNN reported.
The feature, called "AI text classifier," is similar to the plagiarism software Turnitin in that when submitting a body of text, the tool will rate the input on a scale ranging from "likely generated by AI" to "very unlikely."
While educators have been longing for such a tool to combat the increasing use of ChatGPT, OpenAI has admitted that the new feature is "imperfect" and should be "taken with a grain of salt," CNN reported.
"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," Lama Ahmad, policy research director at OpenAI, told the outlet. "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."
Related: Princeton Student Builds ChatGPT Detection App to Fight AI Plagiarism
Despite the imperfection of the new feature, OpenAI told CNN that the decision to release the "AI text classifier" has to do with hopefully deterring individuals from claiming AI text was composed by a human, as well as addressing the question of whether humans have a right to know if they are interacting with artificial intelligence.
"This question is much bigger than what we are doing here; society as a whole has to grapple with that question," Jan Leike, a lead on the OpenAI alignment team, told CNN.
