I have many thoughts… no-mouth-must-scream

    • The_hypnic_jerk [he/him]@hexbear.net
      link
      fedilink
      English
      arrow-up
      34
      ·
      edit-2
      9 days ago

      I read about it a bit ago, I don’t think it’s fake but it’s not “ai-generated”. Basically the model is given a bunch of data, and due to it not having like a humans bias of what a virus is supposed to look like to be stable, it comes up with bizarre variations of an existing strain that wouldn’t likely be thought up of without being able to look at, and process, all the data at the same time. Some of those variations ended up being stable, which surprised the researchers.

      This is my recollection of a sub stack on it I read a week or two ago.

      This also isn’t that surprising, this is the thing models are good at. Analyzing raw data.

      • peeonyou [he/him]@hexbear.net
        link
        fedilink
        English
        arrow-up
        10
        ·
        9 days ago

        it is kinda what they do right? take a bunch of data, and throw shit at the wall to see what sticks statistics-wise. it shouldn’t be THAT surprising that some of the viruses ended up being stable