Info

This event will be held in English.

AI systems are becoming more capable, more autonomous and more deeply embedded in everyday life. As agentic AI systems begin to plan, act and make decisions on behalf of people and organisations, new questions arise: How can such systems be designed to be safe, trustworthy and accountable? What forms of oversight are needed? And how can technical progress be aligned with societal expectations, regulation and public trust?

This IJCAI-ECAI 2026 panel brings together selected experts to discuss new directions for safe AI agents. The discussion will address emerging challenges in agentic AI, including reliability, transparency, human oversight, governance and real-world deployment. The session will include moderated discussion and audience Q&A.

The panel is connected to the Safe Agentic AI Framework and Ecosystem Roadmapping (SAFER) Workshop, which brings together researchers, practitioners, policymakers and students to develop a roadmap for trustworthy agentic AI.

Die Teilnehmenden erklären sich damit einverstanden, dass die im Zusammenhang mit der Teilnahme an der Veranstaltung gemachten Fotos online und in Printmedien ohne Anspruch auf Vergütung verbreitet und veröffentlicht werden können. Sollten Bilder nicht erwünscht sein, muss der Veranstalter explizit darauf hingewiesen werden. / Participants agree that photos taken in connection with their participation in the event may be distributed and published online and in print media without any entitlement to compensation. If participants do not wish to be photographed, they must inform the organizer.

Zur Anmeldung »