Uncovering Vulnerabilities of LLM-Assisted Cyber Threat Intelligence

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Meng, Yuqiao, Tang, Luoxi, Yu, Feiyang, Jia, Jinyuan, Yan, Guanhua, Yang, Ping, Xi, Zhaohan
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916056763727872
author Meng, Yuqiao
Tang, Luoxi
Yu, Feiyang
Jia, Jinyuan
Yan, Guanhua
Yang, Ping
Xi, Zhaohan
author_facet Meng, Yuqiao
Tang, Luoxi
Yu, Feiyang
Jia, Jinyuan
Yan, Guanhua
Yang, Ping
Xi, Zhaohan
contents Large language models (LLMs) are increasingly used to help security analysts manage the surge of cyber threats, automating tasks from vulnerability assessment to incident response. Yet in operational CTI workflows, reliability gaps remain substantial. Existing explanations often point to generic model issues (e.g., hallucination), but we argue the dominant bottleneck is the threat landscape itself: CTI is heterogeneous, volatile, and fragmented. Under these conditions, evidence is intertwined, crowdsourced, and temporally unstable, which are properties that standard LLM-based studies rarely capture. In this paper, we present a comprehensive empirical study of LLM vulnerabilities in CTI reasoning. We introduce a human-in-the-loop categorization framework that robustly labels failure modes across the CTI lifecycle, avoiding the brittleness of automated "LLM-as-a-judge" pipelines. We identify three domain-specific cognitive failures: spurious correlations from superficial metadata, contradictory knowledge from conflicting sources, and constrained generalization to emerging threats. We validate these mechanisms via causal interventions and show that targeted defenses reduce failure rates significantly. Together, these results offer a concrete roadmap for building resilient, domain-aware CTI agents.
format Preprint
id arxiv_https___arxiv_org_abs_2509_23573
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Uncovering Vulnerabilities of LLM-Assisted Cyber Threat Intelligence
Meng, Yuqiao
Tang, Luoxi
Yu, Feiyang
Jia, Jinyuan
Yan, Guanhua
Yang, Ping
Xi, Zhaohan
Cryptography and Security
Artificial Intelligence
Large language models (LLMs) are increasingly used to help security analysts manage the surge of cyber threats, automating tasks from vulnerability assessment to incident response. Yet in operational CTI workflows, reliability gaps remain substantial. Existing explanations often point to generic model issues (e.g., hallucination), but we argue the dominant bottleneck is the threat landscape itself: CTI is heterogeneous, volatile, and fragmented. Under these conditions, evidence is intertwined, crowdsourced, and temporally unstable, which are properties that standard LLM-based studies rarely capture. In this paper, we present a comprehensive empirical study of LLM vulnerabilities in CTI reasoning. We introduce a human-in-the-loop categorization framework that robustly labels failure modes across the CTI lifecycle, avoiding the brittleness of automated "LLM-as-a-judge" pipelines. We identify three domain-specific cognitive failures: spurious correlations from superficial metadata, contradictory knowledge from conflicting sources, and constrained generalization to emerging threats. We validate these mechanisms via causal interventions and show that targeted defenses reduce failure rates significantly. Together, these results offer a concrete roadmap for building resilient, domain-aware CTI agents.
title Uncovering Vulnerabilities of LLM-Assisted Cyber Threat Intelligence
topic Cryptography and Security
Artificial Intelligence
url https://arxiv.org/abs/2509.23573