Higher level overview of the current AI Safety landscape:

13 - First Principles of AGI Safety with Richard Ngo

How Does Claude 4 Think? — Sholto Douglas & Trenton Bricken

Why AI alignment could be hard with modern deep learning

Introduction to various sub-fields in AI Safety:

• Reinforcement Learning from Human Feedback

The True Story of How GPT-2 Became Maximally Lewd

• Scalable Oversight

An overview of 11 proposals for building safe advanced AI — AI Alignment Forum

• Unlearning and Control

Adversarial Machine Learning explained! | With examples.

AI Control: Improving Safety Despite Intentional Subversion — AI Alignment Forum

• Mechanistic Interpretability:

Zoom In: An Introduction to Circuits