Explorar

Noticias de IA

1057 elementos — filtrados, clasificados y sin duplicados

TechCrunch AI Security & Safety
Woman claims her stepfather used Grok to transform childhood photo into explicit imagery
arXiv cs.AI Security & Safety
TopoIntent: Compiling Security Intent into Executable, Compliance-Checked Network Topolog…
arXiv cs.AI Security & Safety
Security and Detectability Analysis of Unicode Text Watermarking Methods against Large La…
arXiv cs.AI Security & Safety
Operationalizing Cyber Threat Intelligence with GraphRAG
arXiv cs.AI Security & Safety
Beyond Handcrafted Security: Towards Self-Evolving Defense for LLM Agents
Bloomberg Technology Security & Safety
What a Killer Drone Swarm Looks Like Up Close
Mashable Security & Safety
Did AI play a role in a grisly double murder?
Wired AI Security & Safety
The Safety Reckoning Inside OpenAI
Hacker News (AI filter) Security & Safety
Text AI watermarks will always be trivial to remove
Ars Technica AI Security & Safety
Claude's new Scarlet Letter watermark is invisible — for now
arXiv cs.AI Security & Safety
BrowseSafe: Understanding and Preventing Prompt Injection Within AI Browser Agents
arXiv cs.AI Security & Safety
Enhancing Linux Privilege Escalation Attack Capabilities of Local LLM Agents
arXiv cs.AI Security & Safety
SegPAR: Class-Centric Decision-Based Sparse Attack for Semantic Segmentation
arXiv cs.AI Security & Safety
Convergent Detour Hijacking: Task-Preserving Resource Amplification in Skill-Based LLM Ag…
arXiv cs.AI Security & Safety
Evaluating LLM Generated Detection Rules in Cybersecurity
Hugging Face Daily Papers Security & Safety
PIPES: Securing Agent Perception with Provenance and Priors
Engadget Security & Safety
How to stop Twitch from training AI on your streams
Ars Technica AI Security & Safety
The web’s newest weapon against AI scrapers is a font
Ars Technica AI Security & Safety
Terabytes of credentials leaked in massive supply-chain attack
Ars Technica AI Security & Safety
Twitch content has trained Amazon AI for years, but users can opt out now
Engadget Security & Safety
German nonprofit files criminal complaint over Meta smart glasses privacy
Hacker News (AI filter) Security & Safety
Someone is running mass vulnerability scans, spoofing AI bots like ClaudeBot
Hugging Face Daily Papers Security & Safety
ToolHazard: Scaling Adversarial Environments for Security Evaluation and Alignment of LLM…
Hugging Face Daily Papers Security & Safety
Towards Model-based Run-time Cybersecurity: On Control-Flow Anomaly Detection, Attack Ide…
Hugging Face Daily Papers Security & Safety
High-Order Liquid Evidence Encoding for Gradual GNSS Spoofing Detection in Autonomous Dri…
Latent Space (swyx) Security & Safety
[AINews] How to steal a Reasoning Trace
Hacker News (AI filter) Security & Safety
Company Offering '100% Human-Written, Never AI' Medical Research Is 100% AI
Bloomberg Technology Security & Safety
'Brave New World' With AI Tied To Hacks: Ledger's Rogers
Bloomberg Technology Security & Safety
Bitcoin 'A Canary In The Coal Mine' For Hacks: Rubenfeld
Bloomberg Technology Security & Safety
Crypto Hacks Easier With AI: Ledger's Rogers