3 ms·
>Not sure I follow why "blaming" someone is necessary to move forward with recovering from a failure. I wasn't clear, but to get that AI to admit that it has m
by 1970-01-01 1y ago
>Not sure I follow why "blaming" someone is necessary to move forward with recovering from a failure.
I wasn't clear, but to get that AI to admit that it has made an error and getting it to actually correct its error is like trying to put a round peg into a square hole. It will take the blame and continue as if nothing needs to change, no matter what prompts you send it.
- mym1990 1y agoI think this is assuming that the model doesn't get better through additional training or feedback, but it seems to me that reward/punishment is actually what drives better prediction over the medium/long run. Right now we are definitely in a place where the models are over confident but I think with time the self correction will become very good.