Artificial Intelligence Policy Tracker
Search
Menu

AI governance glossary

OWASP Top 10 for LLM Applications

What is the OWASP Top 10 for LLM Applications?

The OWASP Top 10 for LLM Applications is the Open Worldwide Application Security Project's list of the ten most critical security risks in applications built on large language models. The 2025 edition runs from LLM01 Prompt Injection to LLM10 Unbounded Consumption.

Why it matters

  • It is the common checklist for LLM security reviews, penetration tests and vendor security questionnaires.
  • Several entries correspond to legal duties: the EU AI Act requires high-risk AI systems to resist attempts to manipulate them, naming data poisoning, model poisoning and adversarial inputs (Article 15).
  • It changes as attacks change: the 2025 edition added system prompt leakage and vector and embedding weaknesses.

In detail

A community-maintained awareness list, first published in 2023 and revised for 2025. The 2025 entries are: prompt injection, sensitive information disclosure, supply chain, data and model poisoning, improper output handling, excessive agency, system prompt leakage, vector and embedding weaknesses, misinformation, and unbounded consumption. Each entry describes the weakness, example attack scenarios and mitigations.

Source: OWASP Top 10 for LLM Applications 2025. A plain-language explanation for orientation, not the legal text; follow the source for the binding wording.

AI red team test plan →

  • Prompt injectionPrompt injection is an attack in which crafted input, typed by a user or hidden in content the model reads, ov...
  • Agentic AIAgentic AI is an AI system that pursues a goal by planning and taking actions with some autonomy, such as call...
  • Google Secure AI Framework (SAIF)Google's Secure AI Framework (SAIF) is a conceptual framework, published by Google in June 2023, for securing...
  • Data poisoningData poisoning is an attack that corrupts an AI model by inserting or altering its training or fine-tuning dat...
  • AI hallucination (confabulation)An AI hallucination is output from a generative AI system that is presented confidently but is false, fabricat...

All glossary terms · Think a definition is off? Tell us; corrections are logged on the corrections page.

Frequently asked questions

What is the OWASP Top 10 for LLM Applications?

The OWASP Top 10 for LLM Applications is the Open Worldwide Application Security Project's list of the ten most critical security risks in applications built on large language models. The 2025 edition runs from LLM01 Prompt Injection to LLM10 Unbounded Consumption.

Where does the term OWASP Top 10 for LLM Applications come from?

This explanation follows OWASP Top 10 for LLM Applications 2025. It is a plain-language paraphrase for orientation; the source has the binding wording.