Not a good look for Mastodon - what can be done to automate the removal of CSAM?

  • HughJanus@lemmy.ml
    link
    fedilink
    arrow-up
    41
    arrow-down
    3
    ·
    edit-2
    1 year ago

    “We got more photoDNA hits in a two-day period than we’ve probably had in the entire history of our organization of doing any kind of social media analysis, and it’s not even close,

    How do you have “probably” and “it’s not even close” in the same sentence?

    Here’s the thing, and what I’ve been saying for a long time about The Fediverse:

    I don’t care what platform you have, if it is sufficiently popular, you’re GOING to have CSAM. You’re going to have alt-right assholes. You’re going to have transphobia, you’re going to have racism and every other kind of discrimination.

    People point fingers at Meta for “allowing” this but there’s no amount of money that can reasonably moderate 3 b-b-billion users. Meta, and probably every other platform that’s not Twitter or False social, does what they can about this.

    Masto and Fedi admins need to be cognizant of the amount of users on their instances and need to have a sufficient number of moderators to manage those users. If they don’t have them, they need to close registrations.

    But ultimately the Fediverse can also create safe-havens for these sorts of things. Making it easy to set up a discriminatory network that has no outside moderation. This is the downside of free speech.

    • Grimpen@lemmy.ca
      link
      fedilink
      arrow-up
      8
      ·
      1 year ago

      Heck, Truth Social uses Mastodon, IIRC.

      Ultimately, it’s software. Even if my home instance does a good job of enforcing it’s CoC, and every instance it federated with does as well, someone else can spin up their own instance, load up on whatever, and I’ll never know or even be aware if it’s never federated with my instance.

    • clutchmatic@lemmy.world
      link
      fedilink
      arrow-up
      4
      arrow-down
      1
      ·
      1 year ago

      People point fingers at Meta for “allowing” this but there’s no amount of money that can reasonably moderate 3 b-b-billion users.

      This is a prime use case for AI technology

      • Dubious_Fart@lemmy.ml
        link
        fedilink
        arrow-up
        3
        ·
        1 year ago

        No, cause then you end up with a case like the guy who lost 15+ years of emails, his phone number, all his photos, his contacts, and everything else he had tied to a google account, because Googles automated detection triggered on a naked photo of their baby, that they sent to the doctor during covid, that the doctor requested, about a rash on the babies diaper area… and no amount of common sense would stay their hand or reverse their ignorant judgement that this man was a child pornographer, and even called the police on him.

    • Dubious_Fart@lemmy.ml
      link
      fedilink
      arrow-up
      2
      arrow-down
      1
      ·
      1 year ago

      Yes, it will be an issue on any platform.

      But how that platform deals with/fights it is what makes the platform good or bad.

      Take Facebook for example in horrible… Facebook is rife with the stuff, and it regularly gets reported… and nothing happens. To the point that a Reporter once confronted them about it during an interview, and Facebook proved it did have the capability to contact law enforcement… by calling them on the reporter who showed them the evidence of it on their platform.