Skip to main content

OWASP LLM Top 10

OWASP LLM Top 10

Table of Contents

Quick Answer

The OWASP LLM Top 10 is a practical risk list for applications that use large language models, generative AI, plugins, tools, agents, or AI-connected workflows. Beginners can use it to understand prompt injection, insecure output handling, data exposure, excessive agency, overreliance, supply-chain risk, and other LLM application security concerns.

What is the OWASP LLM Top 10?

The OWASP LLM Top 10 is a security awareness and risk framework for applications that use large language models. It helps developers, learners, product teams, and security teams discuss common LLM application risks in a shared language.

Who Should Use It?

Developers can use it while designing prompts, tools, and workflows. Security teams can use it during threat modeling and review. Product owners can use it to understand approval, privacy, and monitoring needs. Beginners can use it as a learning roadmap before building AI-powered applications.

The 10 LLM Risks Explained

OWASP LLM riskSimple meaningPractical controlRelated page
Prompt InjectionUntrusted text influences behaviorSeparate instructions from dataPrompt Injection
Insecure Output HandlingModel output is trusted unsafelyValidate and sanitize outputLLM Security
Training Data PoisoningBad data affects behaviorCurate and monitor datasetsChecklist
Model Denial of ServiceCost or resource abuseRate limits and quotasChecklist
Supply Chain VulnerabilitiesRisky models or componentsReview providers and dependenciesLLM Security
Sensitive Information DisclosureSecrets or private data leakData minimization and access controlRAG Security
Insecure Plugin DesignTools expose unsafe actionsLeast privilege and approvalAI Agent Security
Excessive AgencyAI can act too autonomouslyHuman approval and scoped permissionsAI Agent Security
OverrelianceUsers trust wrong outputVerification and human reviewAI Security
Model TheftUnauthorized model access or copyingAccess control and monitoringChecklist

Developer Priority Order

Most teams should start with prompt injection, insecure output handling, sensitive information disclosure, tool/plugin permissions, excessive agency, and monitoring. These risks appear early when an LLM app connects to documents, APIs, databases, code, or business workflows.

Beginner Learning Order

  1. Read the AI Security Roadmap.
  2. Understand Prompt Injection.
  3. Learn LLM Security controls.
  4. Study RAG Security if the app uses documents or vector search.
  5. Study AI Agent Security if the app can call tools or take actions.
  6. Review the LLM Security Checklist.

OWASP LLM Top 10 vs NIST AI RMF vs MITRE ATLAS

FrameworkBest forHow to use it
OWASP LLM Top 10LLM application security risk awarenessUse as a practical checklist for app design and review
NIST AI RMFAI risk governance and trustworthinessUse for risk management, measurement, and organizational process
MITRE ATLASAdversary techniques against AI systemsUse for threat-informed review and security testing plans
OWASP Agentic AI guidanceAutonomous agents and tool-using AIUse for agent permissions, approvals, and excessive-agency review

How to Use the List in a Project

Start by mapping the AI data flow: prompts, retrieved content, model outputs, tools, logs, secrets, and human approvals. For each OWASP risk, document whether the project has a control, where that control lives, and how it is tested. Use the LLM Security Checklist to turn the list into practical review steps.

Explore AI Security Topics

FAQs

The OWASP LLM Top 10 is a practical risk list for applications that use large language models, generative AI, plugins, tools, agents, or AI-connected workflows.

Developers, security teams, product owners, students, and AI builders can use it to understand and review common LLM application risks.

Prompt injection is usually the best first topic because it explains how untrusted text can influence an LLM application.

No. OWASP LLM Top 10 focuses on LLM application security risks, while NIST AI RMF is a broader AI risk management framework.

Map your AI data flows, identify where each risk could appear, add controls, and review the LLM Security Checklist.

Sources and further reading