课程: Responsible AI on AWS: Bedrock Guardrails, Amazon Q Security, and SageMaker Clarify

免费学习该课程！

今天就开通帐号，24,700 门业界名师课程任您挑！

AI security architecture

AI security architecture

课程: Responsible AI on AWS: Bedrock Guardrails, Amazon Q Security, and SageMaker Clarify

开始一个月试用 购买团队方案

AI security architecture

“

- [Instructor] Let's talk through AWS AI security architecture. What this diagram represents is a series of security blog posts recommendations from 2023 and 2024, and these represent the recommended security architectures across different kinds of AI implementation scopes. And the framework is developed by AWS security specialists that are working with our largest customers, and it provides a practical approach to secure the AI workloads while maintaining their traditional software engineering best practices and operational effectiveness. So if we take a look at the first topic here, which is the security scoping matrix, there's these five scopes. First one is Scope 1, or consumer apps. The second scope is enterprise apps. The Scope 3 is pre-trained models. Scope 4 is fine-tuned models. And scope 5 is self-trained models. So Scope 5 is probably going to be a very minor amount of the examples, but just for completeness, it's inside of this as well. And the idea here is that there's a…

内容