GPTZero Case Study (Exploring False Positives)
GPTZero Case Study (Exploring False Positives)
gonzoknows.com
GPTZero Case Study (Exploring False Positives)
4 comments
-
-
The issue with any attempt to identify generated text is that anything generated could have been written by a human. There's also the issue with model diversity. Even if a specific model has common patterns that can be identified, there are hundreds of other models that will have different patterns.
Tried GPTZero with some GPT3.5 text and it was very hit or miss.
Missed a ton of actual GPT texts.
This is what the education sector is using
Oh no.