Conversation

Notices

1. Embed this notice
   purple 👊✊💨 (purple@nya.social)'s status on Tuesday, 10-Mar-2026 04:43:54 JST purple 👊✊💨

   in reply to

   • '(vidak) _(:3」∠)_
   @dirtycommo@anticapitalist.party im suffering this in real time.
   • Doughnut Lollipop 【記録係】:blobfoxgooglymlem: likes this.
   • Embed this notice
     '(vidak) _(:3」∠)_ (dirtycommo@anticapitalist.party)'s status on Tuesday, 10-Mar-2026 04:43:56 JST '(vidak) _(:3」∠)_

     I think using LLMs for serious, mission critical applications is very dangerous.

     From my own experiments the chat bot will make incredibly stupid errors, mixed in with one or two insights.

     I think LLMs should be used by humans as a guide to action only.

     The central place that all these enormous capitalist firms have placed LLMs is very scary to me. These firms only want to make profit. Their bots will say anything to keep you hooked into the attention economy.