2 ms·

Looks like they are trying to use AI for automation and it's screwed up. Again. And it will and must happen again and again to get better. Actually they call t

Looks like they are trying to use AI for automation and it's screwed up. Again. And it will and must happen again and again to get better. Actually they call this stressful work and actively explicitly are working towards not getting workers to do this work for their own benefit if you believe HR and the mental health supporters of this policy. I forget the actual term it's something like "Mental Health Ops'. Why should Google make their contactors suffer digging through video chat when they can get an AI to do it for them? This is literally their policy, it's not satire.

anthony_doan 7y ago
> Looks like they are trying to use AI for automation and it's screwed up. Again. And it will and must happen again and again to get better. Excuse me? Why can't they get better data? It seems here they used email spam data to mark chat from stream. Or something similar along the line. If you ever watch Twitch livestream people spam, it's encourage unless it's a oneman spam. Twitch system is much better. Even if it's subjective I'd take that over what Youtube currently have right now. edit: Also they could validate their model better before rolling it out in production and ruining people lives.
megous 7y ago
Yes, but just block further commenting, lol. If your algorithm has bad false positive rate, don't hook it to "delete this person from Google" action, but something less intursive.