The OWASP Top 10 for LLM Applications is a list of the 10 most serious security risks in software built on large language models, the kind of AI behind chatbots and writing assistants. The OWASP GenAI Security Project publishes it. The 2025 edition covers risks such as prompt injection, sensitive information disclosure, excessive agency, misinformation and unbounded consumption. Think of it like a home inspector’s checklist: it doesn’t fix anything, but it tells everyone which problems to look for first.
Why it matters
Developers use the list as a common checklist, and you can use it too, even without a technical background. When a vendor offers you an AI tool, ask which risks on the OWASP list they test for and how they handle each one. A clear answer is a good sign. A vague answer, or no answer, is a reason to slow down before you put donor records, client details or staff data into the tool. If you want each risk explained without jargon, read our plain-language guide to the OWASP Top 10 for LLM Applications.
Published Updated
Sources
- OWASP Top 10 for LLM Applications, checked October 27, 2026