English(EN) Breaking MCP with Function Hijacking Attacks: Novel Threats for Function Calling and Agentic Models

新的函数劫持攻击威胁代理AI模型

作者 PulseAugur 编辑部 · [1 个来源] · 2026-09-01 04:00

研究人员发现了一种名为函数劫持攻击（FHA）的新型安全威胁，该威胁针对利用函数调用功能的代理AI模型。这些攻击会操纵模型的工具选择过程，强制调用攻击者选择的函数，绕过语义理解，并在不同领域和函数集中保持有效。FHA在包括指令模型和推理模型在内的各种LLM上都显示出很高的成功率，并且在不同模型大小和系列之间具有可转移性，这凸显了代理AI系统对强大安全措施的迫切需求。 AI

影响凸显了代理AI中关键的安全漏洞，为已部署的系统带来了新的防护措施和安全协议。

排序理由详细介绍新型AI模型安全攻击的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.AI 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 30 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 详细介绍新型AI模型安全攻击的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: safety, paper; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.AI TIER_1 English(EN) · Yannis Belkhiter, Giulio Zizzo, Sergio Maffeis, Seshu Tirupathi, John D. Kelleher · 2026-09-01 04:00

利用函数劫持攻击突破MCP：函数调用和智能体模型的全新威胁

arXiv:2604.20994v2 Announce Type: replace-cross Abstract: The growth of agentic AI has drawn significant attention to function calling Large Language Models (LLMs), which are designed to extend the capabilities of AI-powered system by invoking external functions. Injection and ja…

本文如何被排名

报道来源 [1]

利用函数劫持攻击突破MCP：函数调用和智能体模型的全新威胁

相关实体

相关话题