Related papers: Cross-Session Threats in AI Agents: Benchmark, Eva…

Agentic AI systems introduce a security surface that is qualitatively different from that of stateless LLMs. They persist memory, invoke external tools, coordinate with peer agents, and operate across sessions, allowing attacks to emerge…

Agentic AI systems, specifically LLM-driven agents that plan, invoke tools, maintain persistent memory, and delegate tasks to peer agents via protocols such as MCP and A2A, introduce a threat surface that differs materially from standalone…

As generative AI (GenAI) agents become more common in enterprise settings, they introduce security challenges that differ significantly from those posed by traditional systems. These agents are not just LLMs; they reason, remember, and act,…

Autonomous AI agents deployed on platforms such as OpenClaw face prompt injection, memory poisoning, supply-chain attacks, and social engineering, yet existing defences address only the platform perimeter, leaving the agent's own threat…

Existing AI agent safety benchmarks focus on generic criminal harm (cybercrime, harassment, weapon synthesis), leaving a systematic blind spot for a distinct and commercially consequential threat category: agents harming their own…

Personal AI agents like OpenClaw run with elevated privileges on users' local machines, where a single successful prompt injection can leak credentials, redirect financial transactions, or destroy files. This threat goes well beyond…

Large language models are increasingly deployed as *deep agents* that plan, maintain persistent state, and invoke external tools, shifting safety failures from unsafe text to unsafe *trajectories*. We introduce **AgentFence**, an…

Autonomous AI agents executing multi-step tool sequences face semantic attacks that manifest in behavioral traces rather than isolated prompts. A critical challenge is cross-attack generalization: can detectors trained on known attack…

Autonomous AI agents are being deployed with filesystem access, email control, and multi-step planning. This thesis contributes to four open problems in AI safety: understanding dangerous internal computations, removing dangerous behaviors…

Retrieval-augmented generation (RAG) systems have become widely used for enhancing large language model capabilities, but they introduce significant security vulnerabilities through prompt injection attacks. We present a comprehensive…

Modern AI agents execute real-world side effects through tool calls such as file operations, shell commands, HTTP requests, and database queries. A single unsafe action, including accidental deletion, credential exposure, or data…

Frontier AI systems are increasingly capable and deployed in high-stakes multi-agent environments. However, existing AI safety benchmarks largely evaluate single agents, leaving multi-agent risks such as coordination failure and conflict…

As Artificial Intelligence (AI) systems increasingly underpin critical applications, from autonomous vehicles to biometric authentication, their vulnerability to transferable attacks presents a growing concern. These attacks, designed to…

An attacker can split a malicious goal into sub-prompts that each look benign on their own and only become harmful in combination. Existing LLM safety benchmarks evaluate prompts one at a time, or across turns of a single chat, and so do…

AI agents powered by large language models (LLMs) are being deployed at scale, yet we lack a systematic understanding of how the choice of backbone LLM affects agent security. The non-deterministic sequential nature of AI agents complicates…

Large language models (LLMs) are increasingly deployed as autonomous agents in offensive cybersecurity. In this paper, we reveal an interesting phenomenon: different agents exhibit distinct attack patterns. Specifically, each agent exhibits…

Code-capable large language model (LLM) agents are increasingly embedded into software engineering workflows where they can read, write, and execute code, raising the stakes of safety-bypass ("jailbreak") attacks beyond text-only settings.…

Cooperative multi-agent methods for embodied AI are almost universally evaluated under idealized communication: zero latency, no packet loss, and unlimited bandwidth. Real-world deployment on robots with wireless links, autonomous vehicles…

AI agents that interact with their environments through tools enable powerful applications, but in high-stakes business settings, unintended actions can cause unacceptable harm, such as privacy breaches and financial loss. Existing…

Recent AI systems combine large language models with tools, external knowledge via retrieval-augmented generation (RAG), and even autonomous multi-agent decision loops. This agentic AI paradigm greatly expands capabilities - but also vastly…