Saved in:
Bibliographic Details
Main Authors: Burger, Christopher, Talley, Karmece, Trotter, Christina
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2512.23587
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911343857106944
author Burger, Christopher
Talley, Karmece
Trotter, Christina
author_facet Burger, Christopher
Talley, Karmece
Trotter, Christina
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2512_23587
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Can AI Recognize Its Own Reflection? Self-Detection Performance of LLMs in Computing Education
Burger, Christopher
Talley, Karmece
Trotter, Christina
Computers and Society
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.
title Can AI Recognize Its Own Reflection? Self-Detection Performance of LLMs in Computing Education
topic Computers and Society
url https://arxiv.org/abs/2512.23587