TEMPO-Diffusion: Temporally Exposed Malicious Poisoning of Diffusion Models
Explorar
Noticias de IA
1359 elementos — filtrados, clasificados y sin duplicados
Adaptive Evaluation of Out-of-Band Defenses Against Prompt Injection in LLM Agents
ShareLock: A Stealthy Multi-Tool Threshold Poisoning Attack Against MCP
Vulnerability of Natural Language Classifiers to Evolutionary Generated Adversarial Text
From Celebrities to Anyone: Characterizing AI Nudification Content, Technology, and Commu…
Fortress and Gatekeeper: Theorizing Transitive Trust in Third-Party Cybersecurity Risk Go…
Inherited Circuits, Learned Semantics: How Fine-Tuning Creates Evasion Vulnerabilities In…
Prompt Injection in Automated R\'esum\'e Screening with Large Language Models: Single and…
What happened after 2k people tried to hack my AI assistant
Anthropic says Alibaba must be punished for largest Claude cloning attack
Russia allegedly used a forensics platform to hack an activist's phone, despite having it…
What Intermediate Layers Know: Detecting Jailbreaks from Entropy Dynamics
What Does It Mean to Break a Distillation Defense?
SoK: AI Secure Code Generation: Progress, Pitfalls, and Paths Forward
A Hybrid CNN-LSTM Intrusion Detection Framework for Cybersecurity in Smart Renewable Ener…
Color Matters: Trigger Color Affects Success in Federated Backdoor Attacks
PVF:Understanding AI Vulnerability Against SDCs
A Marketplace for AI-Generated Adult Content and Deepfakes
Verifiable Manifest Signing and Transparency Enforcement for Secure MCP-Based LLM Pipelin…
Helpful or Harmful? Evaluating LLM-Assisted Vulnerability Patching via a Human Study
Privacy Vulnerabilities of Attention Layers in Tabular Foundation Models and Protection o…
The Unfireable Safety Kernel: Execution-Time AI Alignment for AI Agents and Other Escapab…
Epistemic Bias Injection: Manipulating LLM Opinion via Selective Context Retrieval
Scaling cybercrime disruption through innovation and AI
Anthropic says Alibaba illicitly extracted Claude AI model capabilities
Anthropic Accuses Alibaba of ‘Illicitly’ Accessing AI Models
AI Snitches Get Glitches: Towards Evading Agentic Surveillance
Microsoft Says Copilot AI Helped Knock Down Cybercrime Tools
Fight Against Child Predators Gets Short Shrift While Cases Explode in AI Era
Scammers Are Using AI to Create Fake Auto Loan Documents