Assessing AI detectors in identifying AI-generated code: Implications for education
Educators are increasingly concerned about the usage of Large Language Models (LLMs) such as ChatGPT in programming education, particularly regarding the potential exploitation of imperfections in Artificial Intelligence Generated Content (AIGC) Detectors for academic misconduct.In this paper, we pr...
Saved in:
Main Authors: | , , , , , , , , |
---|---|
Format: | text |
Language: | English |
Published: |
Institutional Knowledge at Singapore Management University
2024
|
Subjects: | |
Online Access: | https://ink.library.smu.edu.sg/sis_research/9244 https://ink.library.smu.edu.sg/context/sis_research/article/10244/viewcontent/3639474.3640068.pdf |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Singapore Management University |
Language: | English |
id |
sg-smu-ink.sis_research-10244 |
---|---|
record_format |
dspace |
spelling |
sg-smu-ink.sis_research-102442024-09-02T06:44:20Z Assessing AI detectors in identifying AI-generated code: Implications for education PAN, Wei Hung CHOK, Ming Jie WONG, Jonathan Leong Shan SHIN, Yung Xin POON, Yeong Shian YANG, Zhou CHONG, Chun Yong David LO, LIM, Mei Kuan Educators are increasingly concerned about the usage of Large Language Models (LLMs) such as ChatGPT in programming education, particularly regarding the potential exploitation of imperfections in Artificial Intelligence Generated Content (AIGC) Detectors for academic misconduct.In this paper, we present an empirical study where the LLM is examined for its attempts to bypass detection by AIGC Detectors. This is achieved by generating code in response to a given question using different variants. We collected a dataset comprising 5,069 samples, with each sample consisting of a textual description of a coding problem and its corresponding human-written Python solution codes. These samples were obtained from various sources, including 80 from Quescol, 3,264 from Kaggle, and 1,725 from Leet-Code. From the dataset, we created 13 sets of code problem variant prompts, which were used to instruct ChatGPT to generate the outputs. Subsequently, we assessed the performance of five AIGC detectors. Our results demonstrate that existing AIGC Detectors perform poorly in distinguishing between human-written code and AI-generated code. 2024-04-01T07:00:00Z text application/pdf https://ink.library.smu.edu.sg/sis_research/9244 https://ink.library.smu.edu.sg/context/sis_research/article/10244/viewcontent/3639474.3640068.pdf http://creativecommons.org/licenses/by-nc-nd/4.0/ Research Collection School Of Computing and Information Systems eng Institutional Knowledge at Singapore Management University Software Engineering Education AI-Generated Code AI-Generated Code Detection Artificial Intelligence and Robotics Software Engineering |
institution |
Singapore Management University |
building |
SMU Libraries |
continent |
Asia |
country |
Singapore Singapore |
content_provider |
SMU Libraries |
collection |
InK@SMU |
language |
English |
topic |
Software Engineering Education AI-Generated Code AI-Generated Code Detection Artificial Intelligence and Robotics Software Engineering |
spellingShingle |
Software Engineering Education AI-Generated Code AI-Generated Code Detection Artificial Intelligence and Robotics Software Engineering PAN, Wei Hung CHOK, Ming Jie WONG, Jonathan Leong Shan SHIN, Yung Xin POON, Yeong Shian YANG, Zhou CHONG, Chun Yong David LO, LIM, Mei Kuan Assessing AI detectors in identifying AI-generated code: Implications for education |
description |
Educators are increasingly concerned about the usage of Large Language Models (LLMs) such as ChatGPT in programming education, particularly regarding the potential exploitation of imperfections in Artificial Intelligence Generated Content (AIGC) Detectors for academic misconduct.In this paper, we present an empirical study where the LLM is examined for its attempts to bypass detection by AIGC Detectors. This is achieved by generating code in response to a given question using different variants. We collected a dataset comprising 5,069 samples, with each sample consisting of a textual description of a coding problem and its corresponding human-written Python solution codes. These samples were obtained from various sources, including 80 from Quescol, 3,264 from Kaggle, and 1,725 from Leet-Code. From the dataset, we created 13 sets of code problem variant prompts, which were used to instruct ChatGPT to generate the outputs. Subsequently, we assessed the performance of five AIGC detectors. Our results demonstrate that existing AIGC Detectors perform poorly in distinguishing between human-written code and AI-generated code. |
format |
text |
author |
PAN, Wei Hung CHOK, Ming Jie WONG, Jonathan Leong Shan SHIN, Yung Xin POON, Yeong Shian YANG, Zhou CHONG, Chun Yong David LO, LIM, Mei Kuan |
author_facet |
PAN, Wei Hung CHOK, Ming Jie WONG, Jonathan Leong Shan SHIN, Yung Xin POON, Yeong Shian YANG, Zhou CHONG, Chun Yong David LO, LIM, Mei Kuan |
author_sort |
PAN, Wei Hung |
title |
Assessing AI detectors in identifying AI-generated code: Implications for education |
title_short |
Assessing AI detectors in identifying AI-generated code: Implications for education |
title_full |
Assessing AI detectors in identifying AI-generated code: Implications for education |
title_fullStr |
Assessing AI detectors in identifying AI-generated code: Implications for education |
title_full_unstemmed |
Assessing AI detectors in identifying AI-generated code: Implications for education |
title_sort |
assessing ai detectors in identifying ai-generated code: implications for education |
publisher |
Institutional Knowledge at Singapore Management University |
publishDate |
2024 |
url |
https://ink.library.smu.edu.sg/sis_research/9244 https://ink.library.smu.edu.sg/context/sis_research/article/10244/viewcontent/3639474.3640068.pdf |
_version_ |
1814047843256631296 |