Conversation

Notices

1. Embed this notice
   Eugene (evil1@hachyderm.io)'s status on Sunday, 01-Oct-2023 12:13:34 JST Eugene

   • Thomas 🔭🕹️

   @thomasfuchs I wouldn’t engineer a bride based on gpt4 output but it it’s quite good at writing Python and JS…entry to mid level software development will be radically different from now on.

   • Embed this notice
     Scott Jenson (scottjenson@social.coop)'s status on Sunday, 01-Oct-2023 12:42:30 JST Scott Jenson

     • Thomas 🔭🕹️

     @thomasfuchs couldn't agree more. As with so many things, it will likely follow Garners hype cycle and eventually be used but in far more modest ways

   • Embed this notice
     Konnor Rogers (konnorrogers@ruby.social)'s status on Sunday, 01-Oct-2023 13:10:49 JST Konnor Rogers

     • Scott Jenson
     • Thomas 🔭🕹️

     @thomasfuchs @scottjenson something I've wondered is how much compute time it takes per request, and as the models get bigger, I imagine compute times get higher as well so the expense is always getting larger

   • Embed this notice
     ADisorderlyFashion (adisorderlyfashion@mas.to)'s status on Sunday, 01-Oct-2023 22:44:04 JST ADisorderlyFashion

     • Thomas 🔭🕹️

     @thomasfuchs We alredy know the flaws though! What the fuck? Why is nobody addressing this?! Could you imagine if during the 90s when internet was first commercially available to you it would randomly output garbage about half the time and tell you to kill yourself if you pressed f12?