Security and Privacy Challenges of Large Language Models: A Survey

By Badhan Chandra Das, M. Amini, Yanzhao Wu. ACM Computing Surveys, 2024

495 citations recorded by the source corpus.

Large language models (LLMs) have demonstrated extraordinary capabilities and contributed to multiple fields, such as generating and summarizing text, language translation, and question-answering. Today, LLMs have become quite popular tools in natural language processing tasks, with the capability to analyze complicated linguistic patterns and provide relevant responses depending on the context. While offering significant advantages, these models are also vulnerable to security and privacy attacks, such as jailbreaking attacks, data poisoning attacks, and personally identifiable information leakage attacks. This survey provides a thorough review of the security and privacy challenges of LLMs, along with the application-based risks in various domains, such as transportation, education, and healthcare. We assess the extent of LLM vulnerabilities, investigate emerging security and privacy attacks against LLMs, and review potential defense mechanisms. Additionally, the survey outlines existing research gaps and highlights future research directions.

Subfields: Artificial Intelligence

View Paper: https://www.semanticscholar.org/paper/74a2ef37466667c843b6322691c49b0475030cb0

Citations in the corpus

• Small Language Models are the Future of Agentic AI (2025, unknown): “LLMs acquire their language understanding capabilities in accordance with scaling laws [15].”
• Tool learning with large language models: a survey (2024, unknown): “Meanwhile, some studies highlight that LLMs can generate biased or harmful content and potentially leak sensitive information [193–195].”
• A Survey on the Memory Mechanism of Large Language Model-based Agents (2024, unknown): “Their memory can record cross-trial information through successful exemplars, which would be used as references on similar trials.”