5 ms·

I agree that it's the ideal use case for the AI, it's just the issue there is that it's an area where you can't afford any false positives or negatives. All hel

by CM30 3y ago

I agree that it's the ideal use case for the AI, it's just the issue there is that it's an area where you can't afford any false positives or negatives. All hell will break loose if sexual abuse content or footage of terrorism/cartel incidents ends up on random people's Facebook timelines, and another kind of hell will break loose if innocent people see their accounts nuked and their details reported to authorities because the system thought something they posted was illegal content. We've already had both things happen in recent times, and it's been a legal and PR nightmare for affected companies. So while using AI would be great to avoid harm to any human moderators, the system would have to be perfect (or nearly perfect) in order to avoid any side effects that would cause serious harm to innocent people and the company/platform in question.

gpderetta 3y ago
It could be an additional line of defence. It would also ideally run on your user agent.
pixl97 3y ago
>where you can't afford any false positives I disagree with that part of the statement. Companies would just defer to the AI rather than human choices and say "Well, computer says no"
- CM30 3y ago
  Sadly you're probably not wrong here. But the consequences for that sort of attitude when it comes to this particular field are extremely high, and I dread to think of the outcomes in said cases.
  - pixl97 3y ago
    Oh, we already predicted the outcome years ago... https://en.wikipedia.org/wiki/Computers_Don%27t_Argue https://en.wikipedia.org/wiki/Computers_Don%27t_Argue
Denote6737 3y ago
So long as we aren't using the AI to punish the posters. Vis-a-vis the google incident last year with the guy loosing his entire account because of telemedicine. I'm okay with false positives. You took down my video of a sphinx cat because your AI fagged it as NSFW. fine. I can't think of a scenario where random people being able to post things getting blocked occasionally for things being incorrectly flagged is a risk. The risk comes when they then use that to ban and delete accounts. If they just lock the account in a non-destructive way, i.e. block new posting/shadow ban. and reverse it on appeal. This is of course excluding the use of such tools in a non-good faith way.

by CM30 3y ago

gpderetta 3y ago

It could be an additional line of defence. It would also ideally run on your user agent.

pixl97 3y ago

>where you can't afford any false positives I disagree with that part of the statement. Companies would just defer to the AI rather than human choices and say "Well, computer says no"

CM30 3y ago
Sadly you're probably not wrong here. But the consequences for that sort of attitude when it comes to this particular field are extremely high, and I dread to think of the outcomes in said cases.
- pixl97 3y ago
  Oh, we already predicted the outcome years ago... https://en.wikipedia.org/wiki/Computers_Don%27t_Argue https://en.wikipedia.org/wiki/Computers_Don%27t_Argue

Denote6737 3y ago

So long as we aren't using the AI to punish the posters. Vis-a-vis the google incident last year with the guy loosing his entire account because of telemedicine. I'm okay with false positives. You took down my video of a sphinx cat because your AI fagged it as NSFW. fine. I can't think of a scenario where random people being able to post things getting blocked occasionally for things being incorrectly flagged is a risk. The risk comes when they then use that to ban and delete accounts. If they just lock the account in a non-destructive way, i.e. block new posting/shadow ban. and reverse it on appeal. This is of course excluding the use of such tools in a non-good faith way.