Talks
Keynotes and invited talks.
Upcoming talks on trustworthy AI. Slides will be posted here after each talk.
15 December 2026
Keynote
Trustworthy AI for Smart Systems: From Explainability and Uncertainty to Verification and Resilience
International Conference on Technological Innovation and Sustainable Engineering in the Digital Era, Tay Do University, Vietnam
As AI becomes embedded in smart infrastructure, electronics, construction, industrial systems and public decision-making, how can we ensure its outputs
are not only intelligent but also understandable, statistically reliable, verifiable, privacy-preserving and resilient to failures or attacks? The talk presents a
practical framework built on four capabilities: explainability, uncertainty awareness, verification and resilience, with case studies from
intrusion detection, structural monitoring, hallucination correction and federated learning.
Slides: in preparation
Date to be announced
Invited talk
Towards Trustworthy Healthcare AI: From Collaborative Learning to Reliable and Secure AI Assistants
Quang-Vinh Dang, British University Vietnam
Building trustworthy AI for healthcare requires addressing challenges across the whole system lifecycle: how sensitive data are used, how institutions collaborate,
how predictions are interpreted, and how AI assistants behave. The talk is organised around four questions. How can clinical models balance predictive performance,
patient privacy, demographic fairness and uncertainty? How can institutions learn collaboratively when their data differ? How can data owners assess information
disclosure and verify that their data have been forgotten? And how can AI assistants produce better-grounded responses and resist malicious instructions?
Biomedical case studies anchor the discussion, and the talk closes with opportunities for interdisciplinary collaboration and validation with healthcare partners.
Slides: in preparation