> For the complete documentation index, see [llms.txt](https://vai-docs.nebra.one/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://vai-docs.nebra.one/verifiable-ai.md).

# Verifiable AI

VAI makes the agent-centric future more accountable and secure. In particular, **AgentNet** enables credibly neutral coordination and provenance.

Similar to HTTPS, VAI provides authentication and data integrity for User-to-Agent and Agent-to-Agent interactions.

### VAI  Architecture

<figure><img src="/files/NksjhNXWR3EwtfRhsStB" alt=""><figcaption></figcaption></figure>

**VAI (**&#x41;gentNe&#x74;**)** uses Trusted Execution Environments (**TEE**s) and public key cryptography to enable lightweight verification of an agent's responses.&#x20;

### Demo

We demonstrate a chatbot that uses **AgentNet** to provide verified LLM queries:

{% embed url="<https://drive.google.com/file/d/14ykmJXdPlJFfW8Xs3P626f-ctfSkc-J8/view?usp=sharing>" %}

Try it yourself! [Request access through our form](https://forms.gle/2dAWxF5b2iZpXkEfA)