aisecurity.ninja: a curated map of AI security
Updated daily
All 1,306 papers →
Latest papers
Trust Me, I'm Your Developer: Self-Issued Authentication in Large Language Models
Privacy, Robustness, and Fairness Trade-offs in Federated Intrusion Detection: Geometric Indistinguishability at the Aggregation Interface
AlcaTRAz - Anchored Tree-Rule Defense Against Jailbreaks
IndicSafeEval: Safety Robustness of Large Language Models under Multilingual Persuasive Jailbreak Attacks
SENTINEL-RL: Offloading Topological Reasoning from LLM Agents in the Security Operations Center
AI security papers
1,306 entriesEvery new arXiv paper on jailbreaks, prompt injection, and agent security, updated daily.
Code and PoCs
29 entriesTools, fuzzers, and proof-of-concept repositories for testing AI systems.
Frameworks and governance
32 entriesGovernment and institutional guidance: NIST, ENISA, MITRE, OWASP, and more.
Courses
18 entriesTraining and certifications in AI and machine-learning security.
Guides and tutorials
12 entriesPractical write-ups on attacks, defenses, and threat modeling.
Communities
5 entriesWhere AI-security practitioners gather.
Links
7 entriesIncident databases, standards, and references that fit nowhere else.
This is a personal project. I started it as my own repository for everything AI security, then made it public because it seemed useful to other researchers. — Luca Sambucci
