all InfoSec news
CodeCloak: A Method for Evaluating and Mitigating Code Leakage by LLM Code Assistants
April 16, 2024, 4:10 a.m. | Amit Finkman, Eden Bar-Kochva, Avishag Shapira, Dudu Mimran, Yuval Elovici, Asaf Shabtai
cs.CR updates on arXiv.org arxiv.org
Abstract: LLM-based code assistants are becoming increasingly popular among developers. These tools help developers improve their coding efficiency and reduce errors by providing real-time suggestions based on the developer's codebase. While beneficial, these tools might inadvertently expose the developer's proprietary code to the code assistant service provider during the development process. In this work, we propose two complementary methods to mitigate the risk of code leakage when using LLM-based code assistants. The first is a technique …
arxiv code codebase coding cs.cl cs.cr cs.lg cs.pl developer developers efficiency errors expose llm popular real tools
More from arxiv.org / cs.CR updates on arXiv.org
Jobs in InfoSec / Cybersecurity
Azure DevSecOps Cloud Engineer II
@ Prudent Technology | McLean, VA, USA
Security Engineer III - Python, AWS
@ JPMorgan Chase & Co. | Bengaluru, Karnataka, India
SOC Analyst (Threat Hunter)
@ NCS | Singapore, Singapore
Managed Services Information Security Manager
@ NTT DATA | Sydney, Australia
Senior Security Engineer (Remote)
@ Mattermost | United Kingdom
Penetration Tester (Part Time & Remote)
@ TestPros | United States - Remote