AgentHijack: Visual Patch Attacks on Multimodal Computer-Use Agents
Explorar
Noticias de IA
1359 elementos — filtrados, clasificados y sin duplicados
Arbitrary Cipher Attacks Against Large Language Models Do Not Require Fine-Tuning
How Fragile Is Safety Alignment at Frontier Scale? A Single-Direction Attack on a 320B MoE
Cyber-Financial Contagion: Modeling the Propagation of an AI Vendor Compromise Through th…
CS-Guard: Benchmarking LLM Guardrails for Code Generation Security
An Experimental Evaluation of Multimodal Prompt Injection Attacks on Agentic AI Frameworks
Black-Box Red Teaming of Agentic AI: A Taxonomy-Driven Framework for Automated Risk Disco…
Trust Me, I'm Your Developer: Self-Issued Authentication in Large Language Models
Learning Intrusion Response Strategies for OT Systems
Six Chinese AI firms accused of aggressively copying US frontier models
I Let an AI Agent Hack All My Gadgets—and I’d Do It Again
Anthropic Worker Resigns, Warns of AI Risks to Humanity
Ledger Hires New Security Chief as Crypto Hacks Hit $1.4 Billion
Man told ChatGPT he was feeling delusional. ChatGPT insisted he was Jesus.
Worried Anthropic researchers warn that AI ‘could kill all humans’
Anthropic researcher quits, says AI could kill us all by the end of the decade
not much happened today
Hardware Trojan Threats to Multi-Chiplet Photonic Neural Network Accelerators
Style Over Substance: Content-Invariant Wrappers Flip LLM Safety-Judge Verdicts
PiMRef: Deducing Ever-evolving Spear-phishing Emails with Knowledge Base Invariants
Bait-and-Recover: Poisoning Internal Refusal Signals to Defend LLMs against White-Box Edi…
Beyond the Verdict: Evidence-Aligned Evaluation of Visual Prompt-Injection Guardrails
A TTP by TTP Approach: Precise Malware Detection via Malicious TTP Recognition
Multimodal Resource-Exhaustion Attacks on Vision-Language Models via Joint Pixel-Prompt O…
Staying on the Attack Path: Structured State for Long-Horizon Automated Penetration Testi…
Towards a Resilience-Theoretic Foundation for Adversarial Robustness in Industrial Contro…
How to Backdoor Image Knowledge Distillation
Evaluating Deep-Search Agents under Hierarchical Web Evidence Poisoning
FATS: A Prompt Injection Attack Utilizing Feign Security Agents with Deceptive Few-shots …
Versioned Transitive Dependency-Closure Binding and Operation-Time Effect Governance for …