Saved in:
Bibliographic Details
Main Authors: Weichert, James, Dimobi, Chinecherem
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2404.11408
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909172739604480
author Weichert, James
Dimobi, Chinecherem
author_facet Weichert, James
Dimobi, Chinecherem
contents As large language models (LLMs) become increasingly commonplace, concern about distinguishing between human and AI text increases as well. The growing power of these models is of particular concern to teachers, who may worry that students will use LLMs to write school assignments. Facing a technology with which they are unfamiliar, teachers may turn to publicly-available AI text detectors. Yet the accuracy of many of these detectors has not been thoroughly verified, posing potential harm to students who are falsely accused of academic dishonesty. In this paper, we evaluate three different AI text detectors-Kirchenbauer et al. watermarks, ZeroGPT, and GPTZero-against human and AI-generated essays. We find that watermarking results in a high false positive rate, and that ZeroGPT has both high false positive and false negative rates. Further, we are able to significantly increase the false negative rate of all detectors by using ChatGPT 3.5 to paraphrase the original AI-generated texts, thereby effectively bypassing the detectors.
format Preprint
id arxiv_https___arxiv_org_abs_2404_11408
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle DUPE: Detection Undermining via Prompt Engineering for Deepfake Text
Weichert, James
Dimobi, Chinecherem
Artificial Intelligence
As large language models (LLMs) become increasingly commonplace, concern about distinguishing between human and AI text increases as well. The growing power of these models is of particular concern to teachers, who may worry that students will use LLMs to write school assignments. Facing a technology with which they are unfamiliar, teachers may turn to publicly-available AI text detectors. Yet the accuracy of many of these detectors has not been thoroughly verified, posing potential harm to students who are falsely accused of academic dishonesty. In this paper, we evaluate three different AI text detectors-Kirchenbauer et al. watermarks, ZeroGPT, and GPTZero-against human and AI-generated essays. We find that watermarking results in a high false positive rate, and that ZeroGPT has both high false positive and false negative rates. Further, we are able to significantly increase the false negative rate of all detectors by using ChatGPT 3.5 to paraphrase the original AI-generated texts, thereby effectively bypassing the detectors.
title DUPE: Detection Undermining via Prompt Engineering for Deepfake Text
topic Artificial Intelligence
url https://arxiv.org/abs/2404.11408