• Kalkaline @leminal.space
    link
    fedilink
    English
    arrow-up
    48
    ·
    edit-2
    10 months ago

    AI can’t be the last word in what gets marked for misconduct etc., however using it as a screening tool for potentially problematic moments in a law enforcement officer’s encounters would be useful. It’s an enormous task to screen through those hours upon hours of video and probably prohibitively expensive for humans to work through.

    • CameronDev@programming.dev
      link
      fedilink
      English
      arrow-up
      10
      ·
      10 months ago

      Need to be certain the false negative rate is near zero though, or important events could be missed, and with AI its nearly impossible to say that with certainty.

        • RaoulDook@lemmy.world
          link
          fedilink
          English
          arrow-up
          4
          ·
          10 months ago

          Yep we have countless stories from all over the place of people trying to get help with crimes that got no help from the police. Over and over I’ve heard people describe how they were robbed and the police don’t put any effort towards catching the perpetrators or returning the property. And that’s far from the worst of it.

          • pearsaltchocolatebar@discuss.online
            link
            fedilink
            English
            arrow-up
            3
            ·
            10 months ago

            I mean, that’s because there’s usually very little evidence to go on after a robbery, unless you have security cameras.

            Most PDs don’t even have the resources to process evidence from things like murders and rapes, so robbery isn’t super high on their priority list.

      • originalucifer@moist.catsweat.com
        link
        fedilink
        arrow-up
        10
        ·
        10 months ago

        so we should just drop the good in pursuit of perfect?

        ai is just an additional tool to be applied based on its efficacy. the better the tooling gets, the more we can trust its results. but no one…

        no one

        is expecting ai to be perfect, and to be trusted 100%.