VulnerAI: GPT Based Web Application Vulnerability Detection
Software vulnerabilities have been a major threat to everyone and have been exploited to gain unauthorized access to the systems. These vulnerabilities directly impact the confidentiality, integrity, and availability of these software systems. Traditional vulnerability detection tools are not effective and are unable to detect new and emerging vulnerabilities. There have been various incidents that were reported due to vulnerabilities in the web application that resulted in huge losses. It is vital to have accurate vulnerability detection tools. In this paper, the authors investigate and present the tool VulnerAI that can be used for vulnerability detection. VulnerAI uses the power of Generative Pre-trained Transformer (GPT) to provide users with the security insights of their code. The authors analyzed vulnerable PHP code snippets using GPT to identify potential security vulnerabilities. The authors also aim to demonstrate that GPT is a satisfactory tool for vulnerability detection, despite its reliability and use in solving many other tasks.
Authors
- Kevser Ovaz Akpinar (ORCID: https://orcid.org/0000-0002-9859-6855)
- Sohan Singh Saimbhi
Institutions
- Rochester Institute of Technology - Dubai (AE)
- Rochester Institute of Technology (US)
Publication Details
- Published
- 2024-10-25
- DOI
- https://doi.org/10.1109/icamac62387.2024.10828788
- Citations
- 1
- Primary Topic
- Web Application Security Vulnerabilities
- Type
- article
- Field-Weighted Citation Impact
- 0.78