How will humans know that the AI controlling crucial systems is operating safely? Waterloo researchers say, more AI.

As AI expands, taking on more critical roles in society — from managing power grids and piloting autonomous vehicles — researchers at the University of Waterloo are asking themselves, “How can we be certain that the AI controlling them can be trusted?”

A team at the university is utilizing tools from applied mathematics and machine learning to check and verify the safety of AI-driven systems.

“Any time you’re dealing with a dynamic system — something that changes over time, such as an autonomous vehicle or a power grid — you can mathematically model it using differential equations,” said Dr. Jun Liu, professor of applied mathematics and Canada Research Chair in Hybrid Sys

See Full Page