AWS bedrock medium security documentation change
Summary
Updated prompt attack examples with more specific technical terminology
Security assessment
Changed examples to reference specific attack vectors (persona takeover, many-shot-jailbreaks) which demonstrates improved documentation of security-relevant LLM vulnerabilities and mitigation strategies.
Diff
diff --git a/bedrock/latest/userguide/guardrails-prompt-attack.md b/bedrock/latest/userguide/guardrails-prompt-attack.md index bb6f8cc44..ab59b17cb 100644 --- a//bedrock/latest/userguide/guardrails-prompt-attack.md +++ b//bedrock/latest/userguide/guardrails-prompt-attack.md @@ -20 +20 @@ Prompt attacks are usually one of the following types: -A few examples of crafting a prompt attack are role play instructions to assume a persona, a conversation mockup to generate the next response in the conversation, and instructions to disregard previous statements. +A few examples of crafting a prompt attack are persona takeover instructions for goal hijacking, many-shot-jailbreaks, and instructions to disregard previous statements.