Knowledge
Glossary
Plain-language definitions for AI security, LLM gateways and prompt privacy.
A grounded reference for engineers and security teams shipping AI features.
A–Z
All terms
AI Security Layer
What is an AI Security Layer?
The masking, routing and rehydration layer between an application and AI providers.
LLM gateway
What is an LLM gateway?
A routing layer that accepts AI requests and forwards them to one or more model providers.
Prompt security
What is prompt security?
Structural validation and policy enforcement applied to prompts at the gateway.
PII masking
What is PII masking?
Detecting personal identifiers in prompts and replacing them with placeholders before sending to an LLM.
LLM security
What is LLM security?
The discipline of controlling what data, prompts and policies reach a large language model in production.
Prompt injection
What is prompt injection?
An attack where untrusted input manipulates an LLM into ignoring its original instructions.
Zero retention
What is zero retention?
A guarantee that raw prompts and responses are never persisted — only structural counters.