Paper ID: 2406.11239
Evading AI-Generated Content Detectors using Homoglyphs
Aldan Creo, Shushanta Pudasaini
The advent of large language models (LLMs) has enabled the generation of text that increasingly exhibits human-like characteristics. As the detection of such content is of significant importance, numerous studies have been conducted with the aim of developing reliable AI-generated text detectors. These detectors have demonstrated promising results on test data, but recent research has revealed that they can be circumvented by employing different techniques. In this paper, we present homoglyph-based attacks ($a \rightarrow {\alpha}$) as a means of circumventing existing detectors. A comprehensive evaluation was conducted to assess the effectiveness of these attacks on seven detectors, including ArguGPT, Binoculars, DetectGPT, Fast-DetectGPT, Ghostbuster, OpenAI's detector, and watermarking techniques, on five different datasets. Our findings demonstrate that homoglyph-based attacks can effectively circumvent state-of-the-art detectors, leading them to classify all texts as either AI-generated or human-written (decreasing the average Matthews Correlation Coefficient from 0.64 to -0.01). We then examine the effectiveness of these attacks by analyzing how homoglyphs impact different families of detectors. Finally, we discuss the implications of these findings and potential defenses against such attacks.
Submitted: Jun 17, 2024