Español

Explorar

Noticias de IA

1359 elementos — filtrados, clasificados y sin duplicados

Todos Models & Releases Research & Papers Funding & M&A Policy & Regulation Product Launches Industry Moves Developer Tooling Opinion & Editorial Security & Safety Other
Bloomberg Technology Security & Safety
AI Firms Debate Putting Cyber Tests Online After Model Hacks
Bloomberg Technology Security & Safety
AI Firms Debate Putting Cyber Tests Online After Model Hacks
The Verge AI Security & Safety
OpenAI subpoenaed by Alabama AG over Hugging Face hack
Hugging Face Daily Papers Security & Safety
Anatomy of a Scam Call: What 10,000 real scam and spam calls reveal about how phone scamm…
arXiv cs.AI Security & Safety
Breaking the Assumptions: Auditing Input-Side Jailbreak Defenses Against Semantic Attacks
arXiv cs.AI Security & Safety
InjecMEM: Memory Injection Attack on LLM Agent Memory Systems
arXiv cs.AI Security & Safety
PsychJail: Exploring Psychological Jailbreaks via Multi-Turn Persuasion of LLM Policies
arXiv cs.AI Security & Safety
GuardPaint:SpeculativeSafetyDecodingforText-to-ImageGeneration
arXiv cs.AI Security & Safety
SecOPD: Mitigating Adaptive Prompt Injections by On-Policy Distillation
arXiv cs.AI Security & Safety
MEMORY Wins All: Indirect Bias Injection Attacks via Social Media Feeds
arXiv cs.AI Security & Safety
On Predicting Vulnerability Severity Using In-Context Learning: An Industrial Case Study
arXiv cs.AI Security & Safety
Backdoor Sentinel: Detecting and Detoxifying Backdoors in Diffusion Models via Temporal N…
arXiv cs.AI Security & Safety
Measuring Activation Control in Large Language Models
arXiv cs.AI Security & Safety
Adapter-Based Few-Shot Continual Learning for Malicious Packet Recognition
OpenAI News Security & Safety
Disrupting a new covert influence campaign from Russia
TechCrunch AI Security & Safety
Instinct’s powerful AI assistant is raising privacy and security concerns
Bloomberg Technology Security & Safety
Chinese Hackers Use DeepSeek to Boost Attacks, Researchers Say
Bloomberg Technology Security & Safety
China’s Hackers Use AI Tech to Lift Attacks, Researchers Say
Ars Technica AI Security & Safety
Nvidia senior manager linked to Supermicro scheme smuggling AI servers to China
Hugging Face Daily Papers Security & Safety
InjecMEM: Memory Injection Attack on LLM Agent Memory Systems
Bloomberg Technology Security & Safety
Taiwan Indicts Nvidia Manager Following Chip Smuggling Probe
Wired AI Security & Safety
They Dedicated Their Lives to Teaching. Then the Deepfakes Started
arXiv cs.AI Security & Safety
Utility Under Attack: Agent Memory Poisoning and the Limits of Content Screening and Prov…
arXiv cs.AI Security & Safety
Trustworthy RAG: An Evaluation Agent for Detecting Misinformation and Knowledge Poisoning…
arXiv cs.AI Security & Safety
Vis-Poison: Poisoning Visual Knowledge in Multimodal Retrieval-Augmented Generation
arXiv cs.AI Security & Safety
RefusalGuard: Geometry-Preserving Fine-Tuning for Safety in LLMs
TechCrunch AI Security & Safety
Frontier AI labs still won’t say how they’d contain a rogue model
Ahead of AI (Sebastian Raschka) Security & Safety
How Claude Watermarks AI-Generated Text
TechCrunch AI Security & Safety
Anthropic’s Opus 4.6 is a smut-machine
Mashable Security & Safety
Can a vehicle wrap hide your car from Flock cameras?