Slide 3 of 27
Part 1 · What Is It?Slide 3
Slide 3 · The Definition
Now the formal definition — you already know what it means.
OWASP LLM05:2025, in their words, then unpacked.
OWASP LLM05:2025 Official Definition

Insufficient validation, sanitization, and handling of the outputs generated by large language models before they are passed downstream to other components and systems.

What “Downstream” Means

The LLM output doesn’t stay inside the model. It goes somewhere — a web browser, a database query, a shell command, an API call. Each of those is a downstream system with its own rules about what’s safe. Sending raw LLM text into any of them without adapting it for that system’s context is the vulnerability.

What “Sanitization” Means Here

It’s context-specific. The same LLM output might need:

HTML encoding if it goes into a web page (< becomes &lt;)
Parameterized queries if it feeds a database
Sandboxed execution if it’s code that gets run
URL validation if it becomes a fetch target

The Simple Test

If your application would sanitize text from a random user before using it, it must also sanitize text from the LLM. The LLM is not a trusted source. It can be manipulated, and it can make mistakes.

← BackNext → Why LLM output is the new untrusted input