Amazon Bedrock Guardrails supports cross-account safeguards with centralized control and management

This article announces the general availability of cross-account safeguards in Amazon Bedrock Guardrails, enabling centralized enforcement of safety controls across multiple AWS accounts.

• Organization-level enforcement applies single guardrail to all entities via policy settings
• Account-level enforcement automatically applies safeguards to all Bedrock model invocations
• Centralized management reduces administrative burden of monitoring individual accounts
• New feature allows including or excluding specific models from enforcement
• Selective or comprehensive content guarding controls for system and user prompts
• Guardrail configurations remain immutable and cannot be modified by member accounts
• Available in all AWS commercial and GovCloud regions where Bedrock Guardrails exists
• Automated Reasoning checks not supported with this capability

Cross-account safeguards provide unified, centralized protection for responsible AI across organizations while maintaining flexibility for account and application-specific controls.