How to Make AI Boldly Say 'I Don't Know': The Secret of Metacognitive Learning
Going beyond AIs that confidently lie, we explore RLMF, a new learning technique that creates AIs capable of accurately recognizing their limitations and admitting when they 'don't know'.