Jailbreaks
Techniques that bypass safety alignment in instruction-tuned models — roleplay exploits, many-shot jailbreaking, and competing objectives attacks against frontier AI systems.
FRAMEWORK INDEX
Categories
LLM Security
Security vulnerabilities in large language model deployments — insecure output handling, excessive agency, model theft, and inference attacks. Covers the full OWASP LLM Top 10.
Prompt Injection
Direct and indirect prompt injection attacks against LLM-powered applications — techniques, real-world exploits, and mitigations. Mapped to MITRE ATLAS AML.T0051 and OWASP LLM01.
First Look: Security
Proactive security assessments of new AI capabilities as they ship. Every feature analysed through MITRE ATLAS and OWASP LLM Top 10 — mapping attack surface before exploitation begins.
Supply Chain
ML supply chain attacks — malicious model weights on HuggingFace, poisoned pip packages, compromised training pipelines. Mapped to MITRE ATLAS AML.T0010 and OWASP LLM05.
◉ AI THREAT BRIEFING
Stay ahead of the threat.
Twice-weekly digest of critical AI security developments — every story mapped to MITRE ATLAS and OWASP LLM Top 10. Free.
No spam. Unsubscribe anytime.