# Output Filtering: definition and example

URL: https://airuntimesecuritytools.com/glossary/output-filtering/
Updated: 2026-11-16

Output filtering inspects what a model returns and can block, mask or rewrite it before a user or another system receives it. It targets leaked personal data, secrets, unsafe content and replies that break policy. It acts after the model has answered, so it does not change what the model was asked.

## Example

Lasso says its gateway mode can block an interaction before a response returns, mask sensitive data in transit, or allow and alert.

## Sources

- [Lasso Security page](https://www.lasso.security/use-cases/ai-gateway-security) (lasso.security), read 2026-11-16

## Related pages

- [Input Filtering](/glossary/input-filtering/)
- [PII Redaction](/glossary/pii-redaction/)
- [Sensitive Information Leakage](/glossary/sensitive-information-leakage/)
- [Runtime Security for RAG Pipelines](/guides/runtime-security-for-rag-pipelines/)
- [All glossary terms](/glossary/)
- [Model Denial of Service](/glossary/model-denial-of-service/)
- [Guides](/guides/)
