AI Threat Alert
Death by a Thousand Prompts: Open Model Vulnerability Analysis
adversarial testing, we measured each model's resilience against single-turn and multi-turn prompt injection and jailbreak attacks. Our findings reveal pervasive vulnerabilities across all tested models, with multi
OpenGuardrails: A Configurable, Unified, and Scalable Guardrails Platform for Large Language Models
safety violations such as harmful or explicit text generation, (2) model-manipulation attacks including prompt injection, jailbreaks, and code-interpreter abuse, and (3) data leakage involving sensitive or private information
ATA: A Neuro-Symbolic Approach to Implement Autonomous and Trustworthy Agents
models, while exhibiting perfect determinism, enhanced stability against input perturbations, and inherent immunity to prompt injection attacks. By generating decisions grounded in symbolic reasoning, ATA offers a practical and controllable
Protect: Towards Robust Guardrailing Stack for Trustworthy Enterprise LLM Systems
extensive, multi-modal dataset covering four safety dimensions: toxicity, sexism, data privacy, and prompt injection. Our teacher-assisted annotation pipeline leverages reasoning and explanation traces to generate high-fidelity, context
"I know it's not right, but that's what it said to do": Investigating Trust in AI Chatbots for Cybersecurity Policy
chatbots are an emerging security attack vector, vulnerable to threats such as prompt injection, and rogue chatbot creation. When deployed in domains such as corporate security policy, they could
AdvEvo-MARL: Shaping Internalized Safety through Adversarial Co-Evolution in Multi-Agent Reinforcement Learning
role coordination, but their openness and interaction complexity also expose them to jailbreak, prompt-injection, and adversarial collaboration. Existing defenses fall into two lines: (i) self-verification that asks each
Fairness Testing in Retrieval-Augmented Generation: How Small Perturbations Reveal Bias in Small Language Models
concerns regarding security and fairness. Beyond known attack vectors such as data poisoning and prompt injection, LLMs are also vulnerable to fairness bugs. These refer to unintended behaviors influenced
How Diffusion Models Memorize
under memorization due to classifier-free guidance amplifying predictions and inducing overestimation; (ii) memorized prompts inject training images into noise predictions, forcing latent trajectories to converge and steering denoising toward
ReliabilityRAG: Effective and Provably Robust Defense for RAG-based Web-Search
documents. These systems, however, remain vulnerable to attacks on the retrieval corpus, such as prompt injection. RAG-based search systems (e.g., Google's Search AI Overview) present an interesting setting
LAAF: Logic-layer Automated Attack Framework A Systematic Red-Teaming Methodology for LPCI Vulnerabilities in Agentic Large Language Model Systems
pipelines, and external tool connectors face a class of attacks - Logic-layer Prompt Control Injection (LPCI) - for which no automated red-teaming instrument existed. We present LAAF (Logic-layer Automated
Uncovering Security Threats and Architecting Defenses in Autonomous Agents: A Case Study of OpenClaw
OpenClaw ecosystem. We systematically investigate its current threat landscape, highlighting critical vulnerabilities such as prompt injection-driven Remote Code Execution (RCE), sequential tool attack chains, context amnesia, and supply chain
Biosecurity-Aware AI: Agentic Risk Auditing of Soft Prompt Attacks on ESM-Based Variant Predictors
GFMs. SAGE functions through an interpretable and automated risk auditing loop. It injects soft prompt perturbations, monitors model behavior across training checkpoints, computes risk metrics such as AUROC and AUPR
Verifiability-First Agents: Provable Observability and Lightweight Audit Agents for Controlling Autonomous LLM Systems
detection under stealthy strategies, and (iii) resilience of verifiability mechanisms to adversarial prompt and persona injection. Our approach shifts the evaluation focus from how likely misalignment is to how quickly
AULLM++: Structural Reasoning with Large Language Models for Micro-Expression Recognition
propose AULLM++, a reasoning-oriented framework leveraging Large Language Models (LLMs), which injects visual features into textual prompts as actionable semantic premises to guide inference. It formulates AU prediction into
BadTemplate: A Training-Free Backdoor Attack via Chat Template Against Large Language Models
chat templates allows an attacker who controls the template to inject arbitrary strings into the system prompt without the user's notice. Building on this, we propose a training-free
Multi-Turn Jailbreaking of Aligned LLMs via Lexical Anchor Tree Search
injection. LATS reformulates jailbreaking as a breadth-first tree search over multi-turn dialogues, where each node incrementally injects missing content words from the attack goal into benign prompts. Evaluations
Synthetic Voices, Real Threats: Evaluating Large Text-to-Speech Models in Generating Harmful Audio
second leverages audio-modality exploits (Read, Spell, Phoneme) that inject harmful content through auxiliary audio channels while maintaining benign textual prompts. Through evaluation across five commercial LALMs-based TTS systems
Evaluating Adversarial Vulnerabilities in Modern Large Language Models
prompted to circumvent their own safety protocols, and 'cross-bypass', where one model generated adversarial prompts to exploit vulnerabilities in the other. Four attack methods were employed - direct injection, role
Large Language Models for Detecting Cyberattacks on Smart Grid Protective Relays
perfect fault detection accuracy. Additional evaluations demonstrate robustness to prompt formulation variations, resilience under combined time-synchronization and false-data injection attacks, and stable performance under realistic measurement noise levels
Exposing Citation Vulnerabilities in Generative Engines
perspectives of citation publishers and the content-injection barrier, defined as the difficulty for attackers to manipulate answers to user prompts by placing malicious content on the web. GEs integrate