I traced $2 billion in nonprofit grants and 45 states of lobbying records to figure out who’s behind the age verification bills. The answer involves a company that profits from your data writing laws that collect more of it.

    • Cethin@lemmy.zip
      link
      fedilink
      English
      arrow-up
      9
      arrow-down
      1
      ·
      20 hours ago

      To all the people saying “but have you checked if it’s wrong”, think about what that implies. Someone can generate way more garbage using an LLM than you can verify. If we need to check it all first, before dismissing it, that means we need to just accept all LLM garbage, because it’s practically impossible to check it all. No, it should be dismissed first, and someone can check it to tell if it’s accurate. I don’t care that it supports your biases. This is much larger than that.

      • misk@piefed.social
        link
        fedilink
        English
        arrow-up
        11
        arrow-down
        3
        ·
        2 days ago

        Looks like I was right. Why repost this? I don’t believe this was ever truly reviewed by a human and Lemmy users won’t do it since most won’t read past the headline.

        • ØR10N5B3LT@midwest.social
          link
          fedilink
          English
          arrow-up
          7
          arrow-down
          3
          ·
          edit-2
          2 days ago

          since you were right, can you review it for us? since we won’t read past the headlines & you pride yourself on being right, might as well dig in, no? let us know what you find in your review! thanks!

          • misk@piefed.social
            link
            fedilink
            English
            arrow-up
            6
            arrow-down
            1
            ·
            2 days ago

            Why would I make this effort if the autor likely didn’t? If they did all the required research they wouldn’t need an LLM in the first place.

        • BrikoX@lemmy.zipOPM
          link
          fedilink
          English
          arrow-up
          3
          arrow-down
          3
          ·
          2 days ago

          Are you disputing any of the listed external sources? Part of it is governments own federal fillings.

          • misk@piefed.social
            link
            fedilink
            English
            arrow-up
            11
            arrow-down
            1
            ·
            2 days ago

            Yes, because those sources weren’t read by a human. They could state one thing while the LLM hallucinates another.

            • BrikoX@lemmy.zipOPM
              link
              fedilink
              English
              arrow-up
              3
              arrow-down
              6
              ·
              2 days ago

              Every factual claim in this repository cites a primary source (IRS filing, Senate disclosure, state database, legislative record, or published reporting) that can be independently verified. The tool does not change whether Meta’s LD-2 filing lists H.R. 3149, whether DCA has an EIN, or whether Stefanski admitted tech funding under oath. The records exist or they don’t.

              Unless you can point to incorrect primary source you are just wrong.

              • misk@piefed.social
                link
                fedilink
                English
                arrow-up
                10
                arrow-down
                2
                ·
                edit-2
                1 day ago

                LLM output isn’t „correct unless proven wrong”, nothing ever is. LLMs aren’t a reliable technology for research and and all it has done so far is that it has lowered the bar on presenting plausible sounding text.

                I’m not after plausible sounding text, I’m after truth, facts and human insights.

                • BrikoX@lemmy.zipOPM
                  link
                  fedilink
                  English
                  arrow-up
                  2
                  arrow-down
                  3
                  ·
                  1 day ago

                  LLMs aren’t a reliable technology for research and and all it has done so far is that it has lowered the bar on presenting plausible sounding text.

                  Don’t disagree. But again, have you checked those human written primary sources and have a particular dispute or are you just blindly saying everything is wrong without even checking the records for yourself?

                  Also as the diclaimer says LLMs weren’t used for research only for grunt work. You seem to only care about complaining without even having any particular issues to complain about.

                  ProPublica, OpenSecrets and US governments own fillings are human records which I checked and I can see they support the conclusion that was written.