The Symposium ‘Governing Agentic AI’ brings together leading scholars and practitioners to explore one of the most pressing challenges in artificial intelligence today: how to effectively govern increasingly autonomous AI systems.


 

Agentic AI systems, characterized by their ability to make independent decisions, plan actions, and execute complex tasks with minimal human intervention, are rapidly transforming multiple sectors. These technologies hold significant promise, including reducing manual and repetitive work, enhancing cybersecurity through adaptive responses, and expanding access to professional services through the real-time management of complex queries.

At the same time, their growing autonomy raises critical governance concerns. Key issues include ensuring sufficient control and accountability, addressing risks such as prompt injection and data poisoning, and understanding emergent behaviors. Broader challenges also involve bias amplification, accuracy and uncertainty, explainability, liability attribution, and sustainability. Beyond these technical risks, agentic AI may also have far-reaching socioeconomic implications, including potential job displacement.


This two-day symposium provides a platform for interdisciplinary dialogue, fostering discussion on legal, technical, and ethical frameworks for governing agentic AI.


See the full programme here.

Register here