Made by OpenAI, the tool determines whether a text is written by Chat GPT or not
OpenAI, the company behind ChatGPT, has just released a new tool that can determine whether a text is written using AI or not. AI writing tools however can still pass through the tool since it is not completely foolproof.
Talking about the tool in a recent blog post, OpenAI said that this is a classifier tool and has the ability to classify between text written by a human and text that is written by an AI text generating tool.
This tool has the ability to detect not just text that is generated via Chat GPT but also from many other popular text generating AI systems.
According to Chat GPT creators Open AI, this classifier tool will be helpful in identifying academic cheating in assignments and exams and also in scenarios where a Chatbot poses as a human. The creators however also said that the tool is still not able to detect all texts written by AI systems but is effectively learning new classifiers between human and AI generated text.
Details suggest that the classifier tool was only able to identify 26% of the text written by AI whereas the remaining went unnoticed. On the other hand, the tool marked around 9% of human written text as AI-generated.
This goes to show that the tool can still make mistakes and isn’t something to entirely relied upon, however it will still act as a sigh of relief for academics worried about AI generated assignments and homework submissions.
“It should not be used as a primary decision-making tool, but instead as a complement to other methods of determining the source of a piece of text.” said OpenAI while talking about the product functionality.
Read more:
Precision Neuroscience Are Designing Brain Implant Safer, Smarter And Reversible
