AI SECURITY CASES × FRAMEWORKS
OWASP LLM Top 10 (2026) · Categories and AI Security Cases
Explore AI security cases, sources, and defensive analysis across 10 populated OWASP LLM Top 10 (2026) categories.
A curated list of the most critical security risks to large language model applications, published by OWASP Foundation.
- LLM01:2026 · Prompt Injection 86 cases
- LLM02:2026 · Sensitive Information Disclosure 82 cases
- LLM03:2026 · Excessive Agency 68 cases
- LLM04:2026 · Supply Chain 55 cases
- LLM05:2026 · Data and Model Poisoning 25 cases
- LLM06:2026 · Unbounded Consumption 9 cases
- LLM07:2026 · Misinformation 7 cases
- LLM08:2026 · Hidden Context Exposure 5 cases
- LLM09:2026 · Vector and Embedding Weaknesses 9 cases
- LLM10:2026 · Improper Output Handling 34 cases
An empty category does not establish safety. A category relation does not prove a defense is deployed or effective.