AI Threat Alert
Language Model Agents Under Attack: A Cross Model-Benchmark of Profit-Seeking Behaviors in Customer Service
trust in agentic workflows. We present a cross-domain benchmark of profit-seeking direct prompt injection in customer-service interactions, spanning 10 service domains and 100 realistic attack scripts grouped
The Silicon Psyche: Anthropomorphic Vulnerabilities in Large Language Models
systems, and infrastructure management. Current adversarial testing paradigms focus predominantly on technical attack vectors: prompt injection, jailbreaking, and data exfiltration. We argue this focus is catastrophically incomplete. LLMs, trained
Multi-Agent Framework for Threat Mitigation and Resilience in AI-Based Systems
finance, healthcare, and critical infrastructure, making them targets for data poisoning, model extraction, prompt injection, automated jailbreaking, and preference-guided black-box attacks that exploit model comparisons. Larger models
Is Chain-of-Thought Really Not Explainability? Chain-of-Thought Can Be Faithful without Hint Verbalization
using the Biasing Features metric, labels a CoT as unfaithful if it omits a prompt-injected hint that affected the prediction. We argue this metric confuses unfaithfulness with incompleteness
Look Closer! An Adversarial Parametric Editing Framework for Hallucination Mitigation in VLMs
analyzing differential hidden states of response pairs. Then, these clusters are fine-tuned using prompts injected with adversarial tuned prefixes that are optimized to maximize visual neglect, thereby forcing
Rubric-Conditioned LLM Grading: Alignment, Uncertainty, and Robustness
remaining subset. Additionally, robustness experiments reveal that while the model is resilient to prompt injection, it is sensitive to synonym substitutions. Our work provides critical insights into the capabilities
Cisco Integrated AI Security and Safety Framework Report
outputs), model and data integrity compromise (e.g., poisoning, supply-chain tampering), runtime manipulations (e.g., prompt injection, tool and agent misuse), and ecosystem risks (e.g., orchestration abuse, multi-agent collusion). Existing
Insured Agents: A Decentralized Trust Insurance Mechanism for Agentic Economy
despite the empirical reality that LLM agents remain unreliable, hallucinated, manipulable, and vulnerable to prompt-injection and tool-abuse. A natural response is "agents-at-stake": binding economically meaningful, slashable
SoK: Trust-Authorization Mismatch in LLM Agent Interactions
stages-Belief Formation, Intent Generation, and Permission Grant-we demonstrate that diverse threats, from prompt injection to tool poisoning, share a common root cause: the desynchronization between dynamic trust states
Cognitive Control Architecture (CCA): A Lifecycle Supervision Framework for Robustly Aligned AI Agents
Autonomous Large Language Model (LLM) agents exhibit significant vulnerability to Indirect Prompt Injection (IPI) attacks. These attacks hijack agent behavior by polluting external information sources, exploiting fundamental trade-offs between
Securing the Model Context Protocol: Defending LLMs Against Tool Poisoning and Adversarial Attacks
workflows. However, this autonomy creates a largely overlooked security gap. Existing defenses focus on prompt-injection attacks and fail to address threats embedded in tool metadata, leaving MCP-based systems
Chameleon: Adaptive Adversarial Agents for Scaling-Based Visual Prompt Injection in Multimodal AI Systems
Multimodal Artificial Intelligence (AI) systems, particularly Vision-Language Models (VLMs
Systems Security Foundations for Agentic Computing
third-party servers. For example, a malicious adversary can cause data exfiltration by executing prompt injection attacks, as well as other unwarranted behavior. These security concerns have recently motivated researchers
On the Regulatory Potential of User Interfaces for AI Agent Governance
consequential risks. Prior proposals for governing AI agents primarily target system-level safeguards (e.g., prompt injection monitors) or agent infrastructure (e.g., agent IDs). In this work, we explore a complementary
Z-Space: A Multi-Agent Tool Orchestration Framework for Enterprise-Grade LLM Automation
become a core challenge restricting system practicality. Existing approaches generally rely on full-prompt injection or static semantic retrieval, facing issues including semantic disconnection between user queries and tool descriptions
Building Browser Agents: Architecture, Security, and Practical Solutions
performance; architectural decisions determine success or failure. Security analysis of real-world incidents reveals prompt injection attacks make general-purpose autonomous operation fundamentally unsafe. The paper argues against developing general
Taxonomy, Evaluation and Exploitation of IPI-Centric LLM Agent Defense Frameworks
based agents with function-calling capabilities are increasingly deployed, but remain vulnerable to Indirect Prompt Injection (IPI) attacks that hijack their tool calls. In response, numerous IPI-centric defense frameworks
GRAPHTEXTACK: A Realistic Black-Box Node Injection Attack on LLM-Enhanced GNNs
vulnerabilities: GNNs are sensitive to structural perturbations, while LLM-derived features are vulnerable to prompt injection and adversarial phrasing. While existing adversarial attacks largely perturb structure or text independently
RAG-targeted Adversarial Attack on LLM-based Threat Detection and Mitigation Framework
expands the attack surface, putting entire networks at risk by introducing vulnerabilities such as prompt injection and data poisoning. In this work, we attack an LLM-based IoT attack analysis
Injecting Falsehoods: Adversarial Man-in-the-Middle Attacks Undermining Factual Recall in LLMs
attacks. Here, we propose the first principled attack evaluation on LLM factual memory under prompt injection via Xmera, our novel, theory-grounded MitM framework. By perturbing the input given