Explore the concept of AI alignment and its potential solutions in this 28-minute video. Delve into inner and outer alignment, analyze the Bing chatbot's behavior, and examine OpenAI's recent blog update. Investigate cognitive architecture, the postmodernist trap, and constitutional AI approaches. Learn about the importance of designing benevolent AI systems and gain insights into the challenges and potential solutions for creating aligned artificial intelligence.
Overview
Syllabus
- The Hilarious Timeline
- Inner Alignment
- Outer Alignment
- Inferences & Rumors about Bing
- OpenAI Blog Update
- Cognitive Architecture
- Postmodernist Trap
- Constitutional AI
- Benevolent By Design
- Conclusion
Taught by
David Shapiro ~ AI