AI Safety Fundamentals: Alignment

83 Episodes

1. Public by Default: How We Manage Information Visibility at Get on Board

   Published: 5/12/2024
2. Writing, Briefly

   Published: 5/12/2024
3. Being the (Pareto) Best in the World

   Published: 5/4/2024
4. How to Succeed as an Early-Stage Researcher: The “Lean Startup” Approach

   Published: 4/23/2024
5. Become a Person who Actually Does Things

   Published: 4/17/2024
6. Planning a High-Impact Career: A Summary of Everything You Need to Know in 7 Points

   Published: 4/16/2024
7. Working in AI Alignment

   Published: 4/14/2024
8. Computing Power and the Governance of AI

   Published: 4/7/2024
9. AI Control: Improving Safety Despite Intentional Subversion

   Published: 4/7/2024
10. Emerging Processes for Frontier AI Safety

    Published: 4/7/2024
11. AI Watermarking Won’t Curb Disinformation

    Published: 4/7/2024
12. Challenges in Evaluating AI Systems

    Published: 4/7/2024
13. Interpretability in the Wild: A Circuit for Indirect Object Identification in GPT-2 Small

    Published: 4/1/2024
14. Towards Monosemanticity: Decomposing Language Models With Dictionary Learning

    Published: 3/31/2024
15. Zoom In: An Introduction to Circuits

    Published: 3/31/2024
16. Weak-To-Strong Generalization: Eliciting Strong Capabilities With Weak Supervision

    Published: 3/26/2024
17. Can We Scale Human Feedback for Complex AI Tasks?

    Published: 3/26/2024
18. Machine Learning for Humans: Supervised Learning

    Published: 5/13/2023
19. Visualizing the Deep Learning Revolution

    Published: 5/13/2023
20. Four Background Claims

    Published: 5/13/2023