3 ms·
Because they're not using humans to do the reviews anymore. They're putting bots on the frontlines, then using humans to (maybe) handle appeals. But those huma
by huffmsa 7y ago
Because they're not using humans to do the reviews anymore. They're putting bots on the frontlines, then using humans to (maybe) handle appeals.
But those humans are probably undertrained, underpaid and have zero context for each of their 1k cases, so they default to accepting the bots judgement. Which in turn feedback to the bot, justifying it's actions.
- codedokode 7y agoIt might depend on popularity and download count. I am sure if there are some issues with Facebook app, they won't have to talk to a bot and they won't be banned in 12 hours.
- huffmsa 7y agoWell of course. But Facebook also has the resources and manpower to review and fix vague issues in a short timespan. And lawyers who'd be happy to bill thousands of hours making a scene.
- segmondy 7y agoDoes matter, even if they used bots, The bot would detect something. The bot would go through some rules to determine what it would flag. The bot should report on it. We detected something suspicious about your ads. We detected that you're abusing the battery life. We detected adult contect. Give them something! Even a clue!
- huffmsa 7y agoThe code is probably in place to do that, but it probably errors out and no one who needs access to those logs has access or knowledge of them, and the humans handling appeals are probably siloed off from the bots and devs, so they think things are fine, and the devs think things are fine, since no one is complaining.