🛡️ AI Security - hello · Scour

ClawGuard: A Runtime Security Framework for Tool-Augmented LLM Agents Against Indirect Prompt Injection 🔒BPF LSM

arxiv.org·14h

RedSOC: Open-source framework to benchmark adversarial attacks on AI-powered SOCs — 100% detection rate across 15 attack scenarios [paper + code] 🧮Intel MKL-DNN

github.com·5d·DEV

AI Agent Security Vulnerabilities: Risks, Attacks, and Protection Strategies 🐛Fuzzing

aichatspot.online·3h·Hacker News

Why Real-Time Prompt Filtering Is Critical for AI Data Security in 2026 💬Prompt Engineering

dev.to·1d·DEV

Introducing Sisyphus: Autonomous Security for Financial AI Infrastructure 💓PHI Accrual

rogo.ai·1d

The Jagged Frontier of AI Security 🔐Capability Systems

tomtunguz.com·3d

Prompt Injection: The Silent Failure Mode of Enterprise AI Security 💬Prompt Engineering

princetonits.com·5d

ETSI EN 304 223 Securing Artificial Intelligence (SAI); Baseline Cyber Security Requirements for AI Models and Systems 🔒Security

etsi.org·5d

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

huggingface.co·6d·Hacker News

BadSkill: Backdoor Attacks on Agent Skills via Model-in-Skill Poisoning ⚓Anchors

arxiv.org·1d

0DIN is open-sourcing AI security and the hard-earned knowledge behind it 📡SonarQube

blog.mozilla.org·5d

Why Operationalizing AI Security Is the Next Great Enterprise Hurdle 🔐Capability Systems

techrepublic.com·6d

AI Integrity: A New Paradigm for Verifiable AI Governance 🎭Program Synthesis

arxiv.org·14h

The Blind Spot of Agent Safety: How Benign User Instructions Expose Critical Vulnerabilities in Computer-Use Agents ⚓Anchors

arxiv.org·14h

The Salami Slicing Threat: Exploiting Cumulative Risks in LLM Systems 💬Prompt Engineering

arxiv.org·14h

PRISM Risk Signal Framework: Hierarchy-Based Red Lines for AI Behavioral Risk 💬Prompt Engineering

arxiv.org·14h

SkillSieve: A Hierarchical Triage Framework for Detecting Malicious AI Agent Skills 💬Prompt Engineering

arxiv.org·5d

ClawLess: A Security Model of AI Agents 🐛Fuzzing

arxiv.org·5d

SkillAttack: Automated Red Teaming of Agent Skills through Attack Path Refinement ⚓Anchors

arxiv.org·6d

Physical Adversarial Attacks on AI Surveillance Systems:Detection, Tracking, and Visible--Infrared Evasion 👁️Computer Vision

arxiv.org·5d

No more posts from hello's subscribed feeds.

Scour all 23983 feeds Learn more about Feeds