The technical challenge

The case for ensuring that powerful AIs are controlled

Create a free account to track your progress and unlock access to the full course content.

Teaching Claude Why (extended research post)

Create a free account to track your progress and unlock access to the full course content.

Safety Cases: How to Justify the Safety of Advanced AI Systems

Create a free account to track your progress and unlock access to the full course content.

The Open Agency Model

Create a free account to track your progress and unlock access to the full course content.

Ngo and Yudkowsky on alignment difficulty

Create a free account to track your progress and unlock access to the full course content.

Automated Alignment Researchers: Using large language models to scale scalable oversight

Create a free account to track your progress and unlock access to the full course content.

Automated Researchers Can Mitigate Well-Characterized Alignment Failures

Create a free account to track your progress and unlock access to the full course content.

An overview of control measures

Create a free account to track your progress and unlock access to the full course content.

Stress Testing Deliberative Alignment for Anti-Scheming Training

Create a free account to track your progress and unlock access to the full course content.

Risk Report: February 2026

Create a free account to track your progress and unlock access to the full course content.

Review of the Anthropic Sabotage Risk Report: Claude Opus 4.6

Create a free account to track your progress and unlock access to the full course content.

Mapping the mind of a large language model

Create a free account to track your progress and unlock access to the full course content.

Circuit Tracing: Revealing Computational Graphs in Language Models

Create a free account to track your progress and unlock access to the full course content.

Natural Language Autoencoders: Turning Claude’s thoughts into text

Create a free account to track your progress and unlock access to the full course content.

A Toy Model of Mechanistic (Un)Faithfulness

Create a free account to track your progress and unlock access to the full course content.

Would This Change Your Answer? Evaluating Explanations of LLM Behavior in the Wild with Counterfactual Experiments

Create a free account to track your progress and unlock access to the full course content.

The Engineer’s Interpretability Sequence

Create a free account to track your progress and unlock access to the full course content.

A bird's eye view of ARC's research

Create a free account to track your progress and unlock access to the full course content.

Conditioning Predictive Models: Risks and Strategies

Create a free account to track your progress and unlock access to the full course content.

Safety from Honesty in a Disinterested AI Predictor

Create a free account to track your progress and unlock access to the full course content.

Towards Guaranteed Safe AI: A Framework for Ensuring Robust and Reliable AI Systems

Create a free account to track your progress and unlock access to the full course content.

AI Safety Atlas

Create a free account to track your progress and unlock access to the full course content.

Introduction to AI Safety, Ethics, and Society

Create a free account to track your progress and unlock access to the full course content.

ARENA: AI Safety Curriculum

Create a free account to track your progress and unlock access to the full course content.