# Glossary of AI native websites and visitor AI

Source: https://tensorpass.app/glossary

Definitions of the terms the site and the docs use, from bring your own AI and the Model Context Protocol to AI wallets and the cost of inference.

- [AI native website](https://tensorpass.app/glossary/ai-native-website): An AI native website answers questions from its own data and serves its search or booking as tools a model calls. Here are its parts and an example.
- [AI wallet](https://tensorpass.app/glossary/ai-wallet): An AI wallet is a browser extension that keeps your AI accounts encrypted on your computer and lends them to the sites you approve. Here is how it works.
- [Bring your own AI](https://tensorpass.app/glossary/bring-your-own-ai): Bring your own AI lets a person run an app's AI features on their own key, credits or plan. Here is how the approaches work and how AI Connect does it.
- [Inference cost](https://tensorpass.app/glossary/inference-cost): Inference cost is what a provider bills to run a model for each answer, per million input and output tokens. Here is the formula with worked examples.
- [Model Context Protocol (MCP)](https://tensorpass.app/glossary/model-context-protocol): The Model Context Protocol is the open standard AI applications use to call outside tools and data. Here is how it works and how AI Connect sites use it.
- [Prompt injection](https://tensorpass.app/glossary/prompt-injection): Prompt injection is text that changes what an AI model does against its developer's instructions. Here is how it works and how AI Connect's guards stop it.