Understanding and Trusting AI Decisions

Recent advances have enabled AI systems to carry out a wide variety of tasks, from medical imaging to question answering rivaling human performance, but AI still has not been able to replace humans in settings where critical decisions need to be made. MIT researcher Hussein Mozannar will join MIT Horizon to discuss human-AI collaboration and explainability. He will explain how humans can examine AI decisions to understand where AI’s weaknesses lie as well as methods that allow AI to give reasoning behind its decisions, allowing humans to better trust and verify AI’s processes. After the talk, Hussein will take questions from the audience.

Register for this MIT Horizon webinar today: https://mit.zoom.us/webinar/register/WN_lXQodwemRJ-m0FYWgvLQaw