AWS has announced a significant expansion in its Audit Manager platform to strengthen governance over generative artificial intelligence applications, now integrating Amazon SageMaker for more thorough and precise monitoring.
In response to the growing need to control and manage generative AI applications, AWS has released a crucial update to its Audit Manager platform. This new version of the best practices framework facilitates evidence collection and enables continuous auditing of compliance with generative AI workloads through 110 standard controls, preconfigured to meet best practice requirements.
Key Improvements
- Advanced integration: The framework now includes Amazon SageMaker as an additional data source, allowing users to exert stricter control over AI workloads on both Amazon Bedrock and SageMaker.
- Customization and flexibility: Controls are organized into domains covering accuracy, privacy, security, and sustainability, enabling organizations to tailor controls to specific country regulations or evolving global standards.
- Streamlined process: To get started, users simply create an assessment in AWS Audit Manager and select the v2 framework for generative AI best practices. The platform automatically collects evidence and generates detailed reports.
This update underscores AWS’s ongoing commitment to innovation in artificial intelligence, making it easier for organizations to adopt responsible and secure practices in the development and deployment of AI models.
The integration of Amazon SageMaker into AWS Audit Manager represents a significant step toward more effective and automated governance of generative AI applications. This not only strengthens organizations’ ability to comply with security and privacy standards but also promotes adaptability to changing regulations and enhances transparency in the lifecycle of AI models. With these enhanced tools, AWS continues to lead the way in responsibly and securely implementing artificial intelligence in critical business scenarios.
