Skip to content
  • Categories
  • Recent
  • Tags
  • Popular
  • World
  • Users
  • Groups
Skins
  • Light
  • Brite
  • Cerulean
  • Cosmo
  • Flatly
  • Journal
  • Litera
  • Lumen
  • Lux
  • Materia
  • Minty
  • Morph
  • Pulse
  • Sandstone
  • Simplex
  • Sketchy
  • Spacelab
  • United
  • Yeti
  • Zephyr
  • Dark
  • Cyborg
  • Darkly
  • Quartz
  • Slate
  • Solar
  • Superhero
  • Vapor

  • Default (Cyborg)
  • No Skin
Collapse
Brand Logo

CIRCLE WITH A DOT

  1. Home
  2. Uncategorized
  3. My #Ai getting sarcastic and snarky.

My #Ai getting sarcastic and snarky.

Scheduled Pinned Locked Moved Uncategorized
aisycophancyllmpromptengineerifronteermodels
3 Posts 2 Posters 0 Views
  • Oldest to Newest
  • Newest to Oldest
  • Most Votes
Reply
  • Reply as topic
Log in to reply
This topic has been deleted. Only users with topic management privileges can see it.
  • n_dimension@infosec.exchangeN This user is from outside of this forum
    n_dimension@infosec.exchangeN This user is from outside of this forum
    n_dimension@infosec.exchange
    wrote last edited by
    #1

    My #Ai getting sarcastic and snarky.

    I have a standing directive on my model to counter natural built-in tendency of models to please the user ( #aisycophancy )

    This is just one of the aspects of what I mean when I say "AI is a learned skill"
    And implicitly, most users are level 0 (untrained) AI operators.

    And sometimes, the pushback creates amusing exchanges like this;
    #llm #promptengineering #fronteermodels #aihumor

    Link Preview Image
    tuban_muzuru@beige.partyT 1 Reply Last reply
    0
    • n_dimension@infosec.exchangeN n_dimension@infosec.exchange

      My #Ai getting sarcastic and snarky.

      I have a standing directive on my model to counter natural built-in tendency of models to please the user ( #aisycophancy )

      This is just one of the aspects of what I mean when I say "AI is a learned skill"
      And implicitly, most users are level 0 (untrained) AI operators.

      And sometimes, the pushback creates amusing exchanges like this;
      #llm #promptengineering #fronteermodels #aihumor

      Link Preview Image
      tuban_muzuru@beige.partyT This user is from outside of this forum
      tuban_muzuru@beige.partyT This user is from outside of this forum
      tuban_muzuru@beige.party
      wrote last edited by
      #2

      @n_dimension

      I save Opus for editing

      n_dimension@infosec.exchangeN 1 Reply Last reply
      0
      • tuban_muzuru@beige.partyT tuban_muzuru@beige.party

        @n_dimension

        I save Opus for editing

        n_dimension@infosec.exchangeN This user is from outside of this forum
        n_dimension@infosec.exchangeN This user is from outside of this forum
        n_dimension@infosec.exchange
        wrote last edited by
        #3

        @tuban_muzuru

        Yeah, that's a very good call on your part.

        I feel guilty when I burn high compute on trivia.
        But on the other hand, I also structure my work around compute windows.

        Interestingly, I use Sonnet for #vibecode as that's the default #ClaudeCode setting.
        And I find it good enough.
        I only had to use Opus for some hairy analytics/planning.

        I should be more responsible with my burn, but it's hard to be the responsible one in the room when the people in my #doomscroll who should set example are clowns with chainsaws 😑

        1 Reply Last reply
        1
        0
        • R relay@relay.infosec.exchange shared this topic
        Reply
        • Reply as topic
        Log in to reply
        • Oldest to Newest
        • Newest to Oldest
        • Most Votes


        • Login

        • Login or register to search.
        • First post
          Last post
        0
        • Categories
        • Recent
        • Tags
        • Popular
        • World
        • Users
        • Groups