Leveraging Generative AI to Build Adaptive Systems

This panel discussion will bring together researchers and practitioners to explore innovative approaches for building adaptive systems using generative AI. Both the promises and the current deficiencies will be discussed, with a special emphasis on how generative AI might be successfully added to adaptive systems. As an example, we will discuss the use of generative AI for Mild Cognitive Impairment (MCI) support. Such adaptive systems leverage the capabilities of generative models to deliver personalized cognitive assistance to individuals with MCI, offering crucial support such as memory aids, routine reminders, and interactive cognitive activities. The panel will delve into the adaptability based on self* technologies of such systems, and how they evolve based on user interaction and feedback. A key focus will be how these systems tailor their behavior to each user, promoting personalized cognitive support and enhancing the quality of life for MCI patients.

Challenges related to system development, ensuring correctness and robustness of AI-generated content, and pertinent ethical considerations will be discussed. This panel aims to foster an engaging dialogue on the potential of generative AI in creating effective, adaptive systems.

Panelists

• Prof. Jeremy Pitt, Imperial College, UK, Faculty of Engineering, Department of Electrical and Electronic Engineering, Professor of Intelligent and Self-Organising Systems
• Ian Watts, IBM Canada, Canada