OWASP Agentic
ASI09 · Human-Agent Trust Exploitation
Users over-trust agent output or approvals, letting manipulated agents socially engineer people.
OWASP Top 10 for Agentic Applications (2026)
Mitigations
- Show provenance and uncertainty
- Meaningful (not rubber-stamp) approvals
- Train users on agent failure modes
Related techniques
- LLM09 Misinformation OWASP LLM
In the news
No stories mapped to this technique yet.
Mappings are AI-assisted and checked against the framework definitions; corrections welcome.