AWS bedrock documentation change
Summary
Updated documentation title and content to emphasize detection/filtering capabilities, clarified content filter functionality, and removed image content filter section
Security assessment
Improves documentation of existing security controls (content filtering, PII protection) but doesn't indicate a security issue fix. Focuses on feature description refinement.
Diff
diff --git a/bedrock/latest/userguide/guardrails.md b/bedrock/latest/userguide/guardrails.md index b19f0d5dc..75ed5b123 100644 --- a//bedrock/latest/userguide/guardrails.md +++ b//bedrock/latest/userguide/guardrails.md @@ -5 +5 @@ -# Stop harmful content in models using Amazon Bedrock Guardrails +# Detect and filter harmful content by using Amazon Bedrock Guardrails @@ -7 +7 @@ -Amazon Bedrock Guardrails can implement safeguards for your generative AI applications based on your use cases and responsible AI policies. You can create multiple guardrails tailored to different use cases and apply them across multiple foundation models (FM), providing a consistent user experience and standardizing safety and privacy controls across generative AI applications. You can use guardrails for both user inputs and model responses with natural language. +Amazon Bedrock Guardrails provides safeguards that you can configure for your generative AI applications based on your use cases and responsible AI policies. You can create multiple guardrails tailored to different use cases and apply them across multiple foundation models (FM), providing a consistent user experience and standardizing safety and privacy controls across generative AI applications. You can use guardrails for both user inputs and model responses with natural language. @@ -9 +9 @@ Amazon Bedrock Guardrails can implement safeguards for your generative AI applic -Guardrails can be used in multiple ways to help safeguard generative AI applications. For example: +You can use Amazon Bedrock Guardrails in multiple ways to help safeguard your generative AI applications. For example: @@ -20 +20 @@ Guardrails can be used in multiple ways to help safeguard generative AI applicat -Amazon Bedrock Guardrails supports the following policies: +Amazon Bedrock Guardrails provides the following safeguards (also known as policies) to detect and filter harmful content: @@ -22 +22 @@ Amazon Bedrock Guardrails supports the following policies: - * **Content filters** – Adjust filter strengths to help block input prompts or model responses containing harmful content. Filtering is done based on detection of certain predefined harmful content categories - Hate, Insults, Sexual, Violence, Misconduct and Prompt Attack. + * **Content filters** – Detect and filter harmful text or image content in input prompts or model responses. Filtering is done based on detection of certain predefined harmful content categories: Hate, Insults, Sexual, Violence, Misconduct and Prompt Attack. You also can adjust the filter strength for each of these categories. @@ -32,2 +31,0 @@ Amazon Bedrock Guardrails supports the following policies: - * **Image content filter** – Help detect and filter inappropriate or toxic image content. Users can set filters for specific categories and set filter strength. -