Everyday AI Podcast – An AI and ChatGPT Podcast

EP 246: No that's not how ChatGPT works. A guide on who to trust around LLMs

Apr 9, 2024
Explore the tangled web of misinformation surrounding ChatGPT and find out who to trust for accurate advice. Discover how to craft effective prompts while understanding the difference between human and AI writing styles. The discussion highlights significant economic investments in AI technology and sheds light on the surprising capabilities of large language models beyond simple tasks. Learn the importance of authentic communication and get tips for mastering generative AI tools as you navigate this evolving landscape.
Ask episode
AI Snips
Chapters
Transcript
Episode notes
ADVICE

Misleading ChatGPT Screenshots

  • Avoid sharing single ChatGPT responses as representative of its capabilities.
  • Outputs depend heavily on prompt quality, not inherent ChatGPT limitations.
INSIGHT

Generative vs. Deterministic

  • ChatGPT is generative, not deterministic, unlike search engines.
  • Expect diverse outputs even from similar inputs due to its non-deterministic nature.
ANECDOTE

Nike Slogan Example

  • Jordan Wilson analyzes a social media post criticizing a ChatGPT-generated Nike slogan.
  • He compares it to a human-written slogan and one generated with proper prompting.
Get the Snipd Podcast app to discover more snips from this episode
Get the app