Proceedings of the ACM Web Conference 2022 · 2022 · 54 citations · 25 references
ComputingComputer ScienceCybersecurityArtificial IntelligenceComputer SecurityMachine LearningDeep LearningOpen-source IntelligenceSoftware Vulnerability DiscoveryDeep Learning For Code SecurityOpen Source Software SecuritySource Code AnalysisSource Code Vulnerability Detection
Github Copilot, trained on billions of lines of public code, has recently become the buzzword in the computer science research and practice community. Although it is designed to help developers implement safe and effective code with powerful intelligence, practitioners and researchers raise concerns about its ethical and security problems, e.g., should the copyleft licensed code be freely leveraged or insecure code be considered for training in the first place? These problems pose a significant impact on Copilot and other similar products that aim to learn knowledge from large-scale open-source code through deep learning models, which are inevitably on the rise with the fast development of artificial intelligence. To mitigate such impacts, we argue that there is a need to invent effective mechanisms for protecting open-source code from being exploited by deep learning models. Here, we design and implement a prototype, CoProtector, which utilizes data poisoning techniques to arm source code repositories for defending against such exploits. Our large-scale experiments empirically show that CoProtector is effective in achieving its purpose, significantly reducing the performance of Copilot-like deep learning models while being able to stably reveal the secretly embedded watermark backdoors.
25
Kishore Papineni, Salim Roukos, Todd J. Ward et al. · 2001 · 21.5K citations · Full text
Targeted Backdoor Attacks on Deep Learning Systems Using Data Poisoning
Xinyun Chen, Chang Liu, Bo Li et al. · arXiv (Cornell University) · 2017 · 1K citations · Full text
Xiaodong Gu, Hongyu Zhang, Sunghun Kim · 2018 · 556 citations