Amazon Bedrock Guardrails now supports cross-account safeguards in general availability, allowing central security teams to enforce AI safety controls across all AWS accounts in an organization from a single management account. The feature blocks up to 88% of harmful multimodal content and filters hallucinated responses. Administrators can configure guardrails at the organization, OU, or individual account level, with multiple guardrails unioned during model inference. Available in all AWS commercial and GovCloud regions where Bedrock Guardrails is supported.
Sort: