This is great! It’s fascinating to see which ones it’s succeeded on vs the one’s I thought it would succeed at. For instance, I thought the following was going to be a piece of cake:
But the answers from the models were no where close.
Have you seen any improvements in the last 8 months?