This AWS workshop focuses on the security risks and vulnerabilities associated with building and deploying a generative AI application. As artificial intelligence evolves, ensuring its security becomes more complex. In this workshop, we will explore AWS security capabilities and best practices to address potential security vulnerabilities in generative AI applications built on Amazon Bedrock, exploring topics such as:

* Use manual and automated testing as part of application evaluations

* Implement near real-time detection and automated response

* Implement controls related to insecure plugins

* Configure Amazon Bedrock guardrails to detect and prevent prompt manipulation and harmful or offensive content

* Implement controls to help prevent sensitive information disclosure

Participants will explore common security risks such as insecure plugins, prompt manipulation, content filtering, and sensitive information exposure. Through a series of hands-on labs, attendees will not only learn about these vulnerabilities but also apply concrete techniques to mitigate them in real-world scenarios.

Attending this workshop requires moderate knowledge of generative AI technologies and security practices. You will need to bring your laptop to the event. We will provide a complementary lab AWS account to all attendees. There is no cost to join this hands-on workshop.

Lunch will be provided by Wiz