Study: Security-aware prompts improve LLM-generated web app security

A new study published on arXiv investigates whether explicitly requesting security best practices improves the output of large language models when generating web applications. The research generated twelve functionally distinct web applications, with six applications having a baseline prompt and the other six having a security-aware prompt. The security-aware variants produced fewer confirmed security findings, with no critical or high issues, compared to the baseline versions. AI

IMPACT This research suggests that explicitly prompting LLMs for security best practices can lead to more secure web application code, potentially improving the overall security posture of AI-generated software.

RANK_REASON The cluster contains a research paper published on arXiv detailing a study on LLM-generated code security. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• alphaXiv
• arXiv
• CatalyzeX
• CORE Recommender
• DagsHub
• Gotit.pub
• Hugging Face
• large language models
• ScienceCast
• Twin-Prompt Study
• web application security

• paper
• safety

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →