Can AI Recognize Its Own Reflection? Self-Detection Performance of LLMs in Computing Education

TL;DR

This paper evaluates the ability of prominent LLMs to detect AI-generated text in computing education, revealing significant instability and susceptibility to deception that undermine their reliability for academic integrity enforcement.

Contribution

It provides an empirical assessment of LLMs' self-detection capabilities in realistic and deceptive scenarios, highlighting current limitations.

Findings

Detection accuracy drops with deceptive prompts

Models often fail to identify human-written text correctly

Deceptive prompts can completely fool detection models

Abstract

The rapid advancement of Large Language Models (LLMs) presents a significant challenge to academic integrity within computing education. As educators seek reliable detection methods, this paper evaluates the capacity of three prominent LLMs (GPT-4, Claude, and Gemini) to identify AI-generated text in computing-specific contexts. We test their performance under both standard and 'deceptive' prompt conditions, where the models were instructed to evade detection. Our findings reveal a significant instability: while default AI-generated text was easily identified, all models struggled to correctly classify human-written work (with error rates up to 32%). Furthermore, the models were highly susceptible to deceptive prompts, with Gemini's output completely fooling GPT-4. Given that simple prompt alterations significantly degrade detection efficacy, our results demonstrate that these LLMs are currently too unreliable for making high-stakes academic misconduct judgments.