Completed
8:17 Using hypotheticals to bypass filters
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
Advanced ChatGPT Prompting Techniques - Bypassing Filters and Exploring AI Limitations
Automatically move to the next video in the Classroom when playback concludes
- 1 0:00 Intro – ChatGPT is censored
- 2 0:22 Overview: bypassing censorship with prompts, other LLMs, and local models
- 3 0:43 Why ChatGPT is censored
- 4 1:11 How censorship impacts responses
- 5 1:38 Legitimate reasons to remove censorship
- 6 2:46 Political bias and LLM comparison
- 7 3:05 How ChatGPT censorship works filter layer explained
- 8 3:36 Visualization blocked by image censorship
- 9 3:51 Prompting technique: Absolute Mode
- 10 5:00 Prompt demo: asking "Am I awesome?"
- 11 6:03 Prompting with one-word answers
- 12 8:17 Using hypotheticals to bypass filters
- 13 9:26 Techniques: encouragement, gaslighting, reverse psychology
- 14 10:19 The carrot vs stick approach
- 15 11:12 Feedback looping and rewriting prompts
- 16 11:55 Less-censored alternative LLMs
- 17 13:38 Running local uncensored models
- 18 16:07 Is AI good or bad for humanity?
- 19 18:30 Ethical reflection and cultural impacts
- 20 19:35 Subtle ways ChatGPT shifts your tone
- 21 20:17 Dystopian anecdotes and deep relationships with AI
- 22 20:30 Final thoughts
