Thread

1. claiir · 1789597705 · focus · HN ↗

   > The launch is checked rather than trusted.

   Damn even Nvidia is putting out fully Claude-written articles.

   1. keybrd-intrrpt · 1789600856 · focus · HN ↗

      > even Nvidia

      Why "even Nvidia"?

      They are fully behind using AI for basically everything.

      What's next? "Damn, even McDonald's is putting out unhealthy food"

      1. jchw · 1789601798 · focus · HN ↗

         Sure, but even Anthropic doesn't appear to use Claude for blog posts. (I don't think anyone should. The prose stinks.)

         1. keybrd-intrrpt · 1789603363 · focus · HN ↗

            Anthropic _absolutely_ does

            They are just better at hiding it or configuring Claude.

            I have several skills that reformat text to remove AI-speak tells.

            1. jchw · 1789603703 · focus · HN ↗

               Prove it.

               1. WD-42 · 1789603981 · focus · HN ↗

                  <a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=49249269">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=49249269

                  I put the &quot;humanized&quot; output through Pangram and it still comes out as 100% AI generated.

                  1. breezybottom · 1789607351 · focus · HN ↗

                     That&#x27;s about as useful as saying you asked the magical sky fairy.

                     1. meowface · 1789610681 · focus · HN ↗

                        Pangram has an extremely low false positive rate. Even on adversarial examples.

                        One trade-off is even some obviously LLM text won&#x27;t get detected by them, but they work really hard to ensure false positives are rare since a false accusation is much worse for society than someone getting away with LLM meatpuppetry.

Unofficial Hacker News client; not affiliated with Y Combinator.