> correcting a few answers also helps the network better determine entire classes of experessions that deserve more definitive & correct answers – that's not su

> correcting a few answers also helps the network better determine entire classes of experessions that deserve more definitive & correct answers – that's not such a big deal, expecially if this constant-human-guided reinforcement-training is well-disclosed. I gather the point is details aren't disclosed. Some kind of update is happening that we aren't told about, and that calls results into question.