AWS prescriptive-guidance documentation change
Summary
Updated image path for security layers diagram and minor text formatting changes in prompt injection security section
Security assessment
Changes involve image path updates and formatting adjustments without introducing new security concepts or addressing vulnerabilities. The content still discusses prompt injection defenses but doesn't indicate any security incident response.
Diff
diff --git a/prescriptive-guidance/latest/llm-prompt-engineering-best-practices/faq.md b/prescriptive-guidance/latest/llm-prompt-engineering-best-practices/faq.md index d8a13a823..62968e3c5 100644 --- a//prescriptive-guidance/latest/llm-prompt-engineering-best-practices/faq.md +++ b//prescriptive-guidance/latest/llm-prompt-engineering-best-practices/faq.md @@ -9 +9 @@ -**Q. What additional security layers should I consider to prevent prompt injection attacks?** +**Q. What additional security layers should I consider to prevent prompt injection attacks?** **A.** The following diagram shows the three main security layers: LLM input, LLM built-in guardrails, and user-introduced guardrails. @@ -11,3 +11 @@ -**A.** The following diagram shows the three main security layers: LLM input, LLM built-in guardrails, and user-introduced guardrails. - - + @@ -27 +25 @@ Your organization should consider implementing security protocols across all lay -**A.** Depending on your organization's constraints and data landscape, prompt security elements can be owned by the data scientist or developer who is working on a specific generative AI use case or by a central generative AI governance team. When you design the MLOps framework for a generative AI solution and release the solution to the production environment, we recommend that you review the AWS blog posts [FMOps/LLMOps: Operationalize generative AI and differences with MLOps](https://aws.amazon.com/blogs/machine-learning/fmops-llmops-operationalize-generative-ai-and-differences-with-mlops/) and [Operationalize LLM Evaluation at Scale using Amazon SageMaker AI Clarify and MLOps services](https://aws.amazon.com/blogs/machine-learning/operationalize-llm-evaluation-at-scale-using-amazon-sagemaker-clarify-and-mlops-services/) as a starting point. Consider introducing security gates to ensure that proper prompt-level security has been added. +**A.** Depending on your organization's constraints and data landscape, prompt security elements can be owned by the data scientist or developer who is working on a specific generative AI use case or by a central generative AI governance team. When you design the MLOps framework for a generative AI solution and release the solution to the production environment, we recommend that you review the AWS blog posts [FMOps/LLMOps: Operationalize generative AI and differences with MLOps](https://aws.amazon.com/blogs/machine-learning/fmops-llmops-operationalize-generative-ai-and-differences-with-mlops/) and [Operationalize LLM Evaluation at Scale using SageMaker AI Clarify and MLOps services](https://aws.amazon.com/blogs/machine-learning/operationalize-llm-evaluation-at-scale-using-amazon-sagemaker-clarify-and-mlops-services/) as a starting point. Consider introducing security gates to ensure that proper prompt-level security has been added.