Generative AI and foundation models have revolutionized cloud-based AI applications. Concurrently, advancements in embedded machine learning and tinyML are enhancing AI capabilities at the network's edge. As both hardware and algorithms become more efficient, these fields are beginning to intersect. The GenAI@Edge symposium on Empowering Generative AI at the Edge addresses the critical challenge of deploying sophisticated models in edge environments, where computational resources are limited and efficiency is paramount. Traditionally, generative AI models, due to their high computational demands, rely on powerful centralized servers. However, the increasing need for real-time, on-device AI solutions in mobile devices, IoT, AR/VR, and other edge applications calls for innovative approaches in model optimization and efficient algorithm design. For the first time, this symposium brings together experts in edge AI, foundation models, and generative AI to foster connections that will pioneer new frontiers in AI. Our symposium will focus on the current status and future outlook for deploying foundation models and generative AI on resource-constrained embedded devices, integrating them within the edge-to-cloud continuum. We aim to bridge this gap by exploring state-of-the-art techniques in novel computing architectures, model compression, and efficient algorithm design customized for edge computing.