Hear why transparency matters and how exposing model reasoning is helping researchers explore interpretability, from @DKundel at @OpenAI.