AWS Bedrock Guardrails adds safeguard tiers with performance and language options

AI Impact Summary

Amazon Bedrock Guardrails has expanded its capabilities with safeguard tiers, offering users greater control over content moderation performance and language support. This enhancement allows for tailored configurations based on application needs, particularly regarding prompt attack detection and denied topic policies. Users can now choose between Standard tier (with enhanced performance and cross-region inference) and Classic tier (supporting English, French, and Spanish), enabling flexibility in adoption and migration strategies.