Amazon Bedrock Guardrails
Managed input and output policies for applications using Amazon Bedrock or the standalone guardrail API.
Published by Amazon Web Services
How to use it
Set policies for selected topics, sensitive information, and other model interaction risks. Guardrails can evaluate application inputs and model responses, including through the separate ApplyGuardrail API where that fits the deployment.
Choose which policies are needed for each workload and check AWS billing for those policy types. Test the behavior against legitimate traffic before configuring production blocking.
- Access
- Paid cloud service