Detecting AI-Generated Source Code in Student Assignments Using Steganographic Watermarks

This paper presents the use of steganography to ensure the integrity and uniqueness of student programming assignments by embedding hidden student identifiers directly into assignment files. The system helps detect plagiarism and the use of AI models without altering visible content or functionality...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:2025 International Conference on Emerging eLearning Technologies and Applications (ICETA) s. 229 - 235
Hlavní autoři: Horvath, Marek, Bubenkova, Lenka, Pietrikova, Emilia, Corba, Matej
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 13.11.2025
Témata:
Plagiarism
Programming
programming assignments
Prototypes
Python
Source coding
Steganography
Syntactics
Testing
text steganography
unethical behavior
Watermarking
White spaces
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:This paper presents the use of steganography to ensure the integrity and uniqueness of student programming assignments by embedding hidden student identifiers directly into assignment files. The system helps detect plagiarism and the use of AI models without altering visible content or functionality. A prototype tool was developed to generate and analyze assignments using multiple steganographic techniques, and the functionality of the proposed methods was evaluated through testing.
DOI:10.1109/ICETA67772.2025.11280274