OpenAI released a ChatGPT Content Detection Tool that can identify text generated by human or artificial intelligence. Let’s dig in more about this ChatGPT Detector.
The AI classifier is a language model that uses a combination of AI-written and human-written text to identify text created by Artificial Intelligence or Automated Software or ChatGPT. It was developed using a dataset of text pairs. The company stated that it uses several providers to address academic dishonesty and automated misinformation campaigns.
OpenAI admits that the detection tool is unreliable for texts below 1,000 characters. Texts written by AI can also be altered to fool the classifier.
OpenAI stated that the classifier is now publicly available to receive users’ feedback on its usefulness.
The company claims that the classifier incorrectly labels human-written texts as AI-generated text 9% of the time. This error didn’t happen in our testing. Practically, the AI classifier was quite useless in evaluating short pieces of writing in its current state. It will ultimately become a cat and mouse game.
“We acknowledge that identifying AI-written texts has been an important topic of discussion among educators. Equally important is recognizing the limits and impacts of AI-generated text classifiers within the classroom.”
ChatGPT launched in November and has enjoyed widespread popularity among millions of users. Yesterday, the chatbot broke one more record of crossing 10 Million daily users a day. However, many of the largest U.S. school districts, including New York City, have since banned the chatbot due to concerns that students might cheat on the text generator or plagiarize.
OpenAI stated that it is in contact with educators to discuss ChatGPT’s limitations and capabilities and will continue its work on detecting AI-generated texts.
Please try this free ChatGPT content detector from here: OpenAI AI Classifier