Changelog Master Feed

Success (and failure) in prompting (Practical AI #213)

7 snips
Feb 28, 2023
Ask episode
AI Snips
Chapters
Transcript
Episode notes
INSIGHT

Understanding AI Model Behavior

  • Generative AI models show both impressive creativity and disturbing failures.
  • Practitioners must understand what leads to good and bad model behavior to use them reliably.
ANECDOTE

AI Enables Endless Creativity

  • Jeremy Fisher created an automated D&D referee that generates infinite unique text and images.
  • This demonstrates the appeal of AI's endless creative abilities in applications like gaming and advertising.
INSIGHT

AI Mirrors Human Biases

  • AI models reflect human biases and negative online sentiments because they train on public data.
  • Disturbing or dark outputs from models actually mirror society's own complex content.
Get the Snipd Podcast app to discover more snips from this episode
Get the app