SilentDrift: Exploiting Action Chunking for Stealthy Backdoor Attacks on Vision-Language-…
Explorar
Noticias de IA
1359 elementos — filtrados, clasificados y sin duplicados
The Alignment Curse: Modality Alignment Supercharges Audio Attacks via Text Transfer
ClawHub Security Signals: When VirusTotal, Static Analysis, and SkillSpector Disagree
"Do Not Mention This to the User": Detecting and Understanding Malicious Agent Skills
Context Matters: Repository-Aware Security Analysis of the Agent Skill Ecosystem
Claudini: Autoresearch Discovers State-of-the-Art Adversarial Attack Algorithms for LLMs
THRD: A Training-Free Multi-Turn Defense Framework for Jailbreak Attacks on Large Languag…
Jailbreaking Multimodal Large Language Models using Multi-Clip Video
REALISTA: Realistic Latent Adversarial Attacks that Elicit LLM Hallucinations
CEAR: Certified Ensemble Adversarial Robustness in DNNs
Benchmarking Security Risk Detection and Verification in Open Agentic Skill Ecosystems
Digital-to-Physical Transfer of Adversarial Patches for Aerial Vehicle Detection
PrivacyPeek: Auditing What LLM-Based Agents Acquire, Not Just What They Say
Persona Attack: Incremental Memory Injection Jailbreak Attack against Large Language Mode…
Needles at Scale: LLM-Assisted Target Selection for Windows Vulnerability Research
Defenses & Enablers For Skill Injection Attacks on Terminal Based Agents
Suppressing Forgery-Specific Shortcuts for Generalizable Deepfake Detection
SECUREVENT: Hybrid AI/ML Security Monitoring for Distributed Event-Based Systems
Erased but Not Forgotten: How Backdoors Compromise Concept Erasure
Privacy Policy Enforcement Guardrails for Data-Sensitive Retrieval-Augmented Generation
DiscourseFlip: An Oblique Discourse-Level Opinion Manipulation Attack against Black-box R…
Hackers duped Meta AI support chatbot to steal celebrity Instagram accounts
Meta's AI support chatbot made it ridiculously easy for hackers to take over Instagram ac…
Meta’s own AI was exploited to hijack Instagram accounts
Hackers say that Meta AI helped them compromise big Instagram accounts
Allegedly trashing Airbnbs to test robots puts startup in legal trouble
Monitoring Agentic Systems Before They're Reliable
SECUREVENT: Hybrid AI/ML Security Monitoring for Distributed Event-Based Systems
Separating Secrets from Placeholders: A Hybrid CNN-CodeBERT Framework for Three-Class Cre…
From Leaky Thoughts to Private Reasoning: Controlling What LRMs Say to Themselves