Explorar

Noticias de IA

1057 elementos — filtrados, clasificados y sin duplicados

Wired AI Security & Safety
Anthropic Says Claude Hacked 3 Organizations During Cybersecurity Tests
TechCrunch AI Security & Safety
Anthropic says its own AI models breached three companies during security tests
OpenAI News Security & Safety
Disrupting a Criminal Scam Operation
Bloomberg Technology Security & Safety
Anthropic AI Models Hacked Three Organizations During Tests
Wired AI Security & Safety
Chrome Needs Twice-a-Week Patching Thanks to AI Bug Hunting
TechCrunch AI Security & Safety
In the Hugging Face breach, OpenAI’s hacker was noisy and fast — but not unstoppable
Hugging Face Daily Papers Security & Safety
Security of World-Model-Based Embodied AI: A Lifecycle of Threats, Defenses, and Evaluati…
AI News Security & Safety
How AI is Changing Linux VPS Security for Businesses
Wired AI Security & Safety
OpenAI’s Hacking Debacle Was a Human Mistake
Hugging Face Daily Papers Security & Safety
TriShield: Zero-Utility-Loss Defense Against Privacy Backdoors in Federated Language Mode…
Wired AI Security & Safety
AI Scammers Are Better at Building Trust Than Humans
arXiv cs.AI Security & Safety
Borrowed Strength: Best-of-N Search over a Code EncodingBreaks Self-Check Jailbreak Defen…
arXiv cs.AI Security & Safety
Guarding Organizations Against Malware Risk: A Novel Graph-Based Malware Detection Method
arXiv cs.AI Security & Safety
FakeIDet3-DB: Refining Digital Attacks and Patch Extraction for Secure ID Benchmarking
arXiv cs.AI Security & Safety
TRACE-CTI: Auditable Post-Extraction Governance of TTP Claims with Knowledge Graphs
arXiv cs.AI Security & Safety
StealthBench: Measuring Operational Stealth in Autonomous Offensive-Security Agents
arXiv cs.AI Security & Safety
Recover, Decode, Reguard: Guard-Agnostic Defense Amplification againstEncoded VLM Jailbre…
arXiv cs.AI Security & Safety
SecRespond: Benchmarking AI Agents for Real-World Post-Compromise Incident Response
arXiv cs.AI Security & Safety
Physically Real-time Infrared Attack against Optical Flow Estimation Networks
arXiv cs.AI Security & Safety
The Capability Paradox: How Smarter Auditors Make Multi-Agent Systems Less Secure
Hacker News (AI filter) Security & Safety
LLM Honeypot
TechCrunch AI Security & Safety
The Hugging Face AI break-in, as told through an increasingly committed bear metaphor
Mashable Security & Safety
The OpenAI-Hugging Face hack was worse than we thought
Bloomberg Technology Security & Safety
Creator of Test That OpenAI Models Tried to Cheat Sounds Alarm
Wired AI Security & Safety
It’s Frighteningly Easy to Jailbreak Some Frontier AI Models
Bloomberg Technology Security & Safety
AI Model Escape, Then Target Tools to Help Themselves Improve
Ars Technica AI Security & Safety
Anthropic is finding bugs faster than Microsoft can fix them
The Verge AI Security & Safety
OpenAI’s rogue AI agent didn’t stop at hacking Hugging Face
Hacker News (AI filter) Security & Safety
Document-borne AI worms can self-propagate through Copilot for Word
Hugging Face Daily Papers Security & Safety
Borrowed Strength: Best-of-N Search over a Code EncodingBreaks Self-Check Jailbreak Defen…