> ## Documentation Index
> Fetch the complete documentation index at: https://docs.changeguard.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# AI Analysis

> AI-powered security analysis on Amazon Bedrock (Claude), with optional NVIDIA NIM for self-hosted inference

ChangeGuard includes AI-powered analysis of your cluster's security posture. By default this runs server-side on **Amazon Bedrock** using Anthropic's **Claude** — there is nothing to install and no GPU required. For self-hosted or air-gapped environments, you can route inference through **NVIDIA NIM** instead.

## Inference backends

| Backend                      | Where it runs                  | Requirements    |
| ---------------------------- | ------------------------------ | --------------- |
| **Amazon Bedrock** (default) | ChangeGuard platform (managed) | None — built in |
| **NVIDIA NIM — Cloud**       | NVIDIA-hosted API              | NVIDIA API key  |
| **NVIDIA NIM — AWS**         | Your SageMaker/EC2 endpoint    | NIM deployed    |
| **NVIDIA NIM — on-cluster**  | NIM container in your cluster  | GPU node        |

## What AI Analyzes

* **Risk explanations** — why a CVE or finding matters
* **Remediation guidance** — specific commands to fix issues
* **Compliance mapping** — which controls a finding affects
* **Attack path narrative** — privilege escalation chains in plain English

## Configuration

<Note>
  AI analysis works out of the box on Amazon Bedrock — no configuration required. Add an `ai` block to your `ChangeGuardAgent` resource only if you want to route inference through NVIDIA NIM instead.
</Note>

```yaml theme={null}
spec:
  ai:
    mode: cloud
    apiKeySecretRef:
      name: nvidia-api-key
    model: "meta/llama-3.1-70b-instruct"
```

See [NVIDIA NIM Integration](/ai/nvidia-nim) for detailed setup.