OWASP LLM
LLM09 · Misinformation
Confident but false outputs (including hallucinated packages or APIs) lead users and systems into unsafe decisions.
OWASP Top 10 for LLM Applications (2025)
Mitigations
- Ground answers in verified sources with citations
- Verify generated dependencies before install
- Communicate uncertainty
Related techniques
- ASI09 Human-Agent Trust Exploitation OWASP Agentic
In the news
-
Measuring Hallucination in Large Language Models for Cyber Threat Intelligence: An Exploratory Study
An exploratory study introduces HalluVision to measure hallucinations in LLM-generated cyber threat intelligence across 4,940 cybersecurity articles.
Tools
Detect or mitigate it
- Giskard LLM red-teaming · Trial
- Guardrails AI Guardrail · Trial
- LangKit Guardrail · Hold
Mappings are AI-assisted and checked against the framework definitions; corrections welcome.