🛡️ AI Security - hop1.ng.1357 · Scour

SkillSieve: A Hierarchical Triage Framework for Detecting Malicious AI Agent Skills 🛡️AI Safety

arxiv.org·1d

PIGuard: Prompt Injection Guardrail via Mitigating Overdefense for Free 💉Prompt Injection

injecguard.github.io·6d·Hacker News

Y2K 2.0: The AI security reckoning 🔓Hacking

anildash.com·15h

Show HN: Prompt injection detector beats ProtectAI by 19% accuracy, 8.9x smaller 💉Prompt Injection

huggingface.co·1d·Hacker News

AI Safety at the Frontier: Paper Highlights of February & March 2026 🛡️AI Safety

lesswrong.com·6d·Hacker News

Safeguarded AI 🛡️AI Safety

aria.org.uk·2d·Hacker News

Show HN: BrokenClaw Part 5: GPT-5.4 Edition (Prompt Injection) 💉Prompt Injection

veganmosfet.codeberg.page·2d·Hacker News

ShieldStack TS – LLM security middleware for TypeScript 💉Prompt Injection

github.com·5d·Hacker News

Ask HN: What's the state of multimodal prompt injection defence in 2026? 💉Prompt Injection

news.ycombinator.com·2d·Hacker News

Scaling Managed Agents: Decoupling the brain from the hands 🔧Agent Tooling

anthropic.com·1d·Hacker News, Hacker News, Hacker News

Your Agent Is Mine: Measuring Malicious Intermediary Attacks on the LLM Supply Chain 💉Prompt Injection

arxiv.org·11h·Hacker News

Securing AI & Agentic Applications 💉Prompt Injection

billdx.github.io·3d·Hacker News

DeepMind's paper on p0wning Claws and what we learned 💉Prompt Injection

suthakamal.substack.com·2d·Substack

How to emotionally grasp the risks of AI Safety 🛡️AI Safety

lesswrong.com·6d·Hacker News

Securing AI infrastructure to prevent backdoors and sabotage 🔎AI Auditing

the-substrate.net·4d·Hacker News

Cryptographic enforcement and semantic monitoring for your AI 🔎AI Auditing

imladri.com·4d·Hacker News

Anthropic expands partnership with Google and Broadcom for multiple gigawatts of next-generation compute 🛡️Anthropic PBC

anthropic.com·3d·Hacker News

ClawLess: A Security Model of AI Agents 🕹️Agentic AI

arxiv.org·1d

🥇Top AI Papers of the Week 🛡️AI Safety

nlp.elvissaravia.com·5d

Physical Adversarial Attacks on AI Surveillance Systems:Detection, Tracking, and Visible--Infrared Evasion 🛡️AI Safety

arxiv.org·1d

Loading more...