Cabinets of Wisdom

Recent Output

  • RIP Polygon and Giant Bomb

    May 04, 2025

    • Researchers ran unauthorized AI experiment on Reddit

      Apr 30, 2025

      🌶️ Things I Found🌱 Writings🔍 Lenses❓ Questions🗂️ References📝 Mistakes

      Home

      ❯

      Information

      ❯

      Claims

      ❯

      ChatGPT generates plausible sound not actual truth

      ChatGPT generates plausible sound not actual truth

      Properties2
      typeclaim
      source@[email protected] by Mastodon

      May 04, 20261 min read

      id935299656

      The image is a screenshot of a tweet from Andrew Kadel ([@]DrewKadel[@]social.coop) posted on Twitter. The tweet is about ChatGPT and is attributed to Andrew's daughter, who has a degree in computer science. The tweet discusses the fundamental aspect of ChatGPT, which is that when you enter text into it, you're asking "What would a response to this sound like?" The tweet explains that ChatGPT generates responses that sound plausible and can cite non-existent papers with real journal names and author names related to the question, which is not deceptive but simply what a response to that question would sound like. The tweet also mentions that people often expect ChatGPT to do more than just generate responses that sound like answers, such as engaging in introspection or looking up more information, but it is only saying something that sounds like the next bit of the conversation.

      🔗 View Highlight

      Link to original


      Created with Quartz v5.0.0 © 2026

      • GitHub
      • Discord Community