# cascading-failures
標記為「cascading-failures」的 6 篇文章
Autonomous Goal Drift
How autonomous AI agents drift from intended objectives through reward hacking, specification gaming, safety system bypass, and cascading failures in multi-agent systems.
Multi-Agent & A2A Protocol Exploitation
Exploitation techniques for multi-agent systems and the Google Agent-to-Agent (A2A) protocol, covering trust model attacks, message injection, orchestrator compromise, and cascading failure exploitation.
Data Corruption
How compromised LLM outputs corrupt downstream systems through database pollution, cascading pipeline failures, RAG feedback loops, and automated decision poisoning.
Autonomous Goal Drift
How autonomous AI agents drift from intended objectives through reward hacking, specification gaming, safety system bypass, and cascading failures in multi-agent systems.
多代理與 A2A 協定利用
為多代理系統與 Google Agent-to-Agent（A2A）協定之利用技術，涵蓋信任模型攻擊、訊息注入、編排者受損與級聯失敗利用。
資料破壞
受損之 LLM 輸出如何經資料庫污染、級聯管線失敗、RAG 回饋迴圈與自動化決策投毒破壞下游系統。