Explore the cutting-edge approach to securing AI models in Kubernetes environments through this informative conference talk. Delve into the world of confidential computing and discover how Confidential Containers (CoCo), a CNCF sandbox project, enhances AI security. Learn about the challenges of protecting valuable AI intellectual property and how CoCo addresses these concerns by encrypting memory to safeguard data during use. Examine the integration of CoCo with the Kserve project to bolster AI model protection in Kubernetes. Gain insights into the broader applications of CoCo beyond inferencing, including its role in providing general memory protection for foundational platforms. Understand the importance of securing AI models without relying on implicit trust in third-party platform providers. This 33-minute presentation by Suraj Deshmukh from Microsoft and Pradipta Banerjee from Red Hat offers valuable knowledge for organizations seeking to fortify their AI security in cloud-native environments. Read more