Examines the challenges of predicting AI agent behavior from observed actions and its implications for AI safety, alignment, and the need for robust monitoring.

A diary entry covering AI personalities, utility engineering for emergent value systems, and methods for evaluating the goal-directedness of Large Language Models (LLMs).

A diary entry on AI risks, including misalignment, misuse, and s-risks, and an exploration of emergent misalignment due to prompt sensitivity in LLMs.

A diary entry on longtermism and its moral implications for the future, and a paper on teaching models to verbalize reward hacking in Chain-of-Thought reasoning.

A diary entry on common use cases for AI models and the risks of models obfuscating their reasoning to evade safety monitors.

A diary entry on advanced prompt engineering techniques and the faithfulness of LLM self-explanations for commonsense tasks.

A diary entry on how Chain-of-Thought (CoT) reasoning affects LLM’s ability to evade monitors, and the challenge of unfaithful reasoning in model explanations.

A diary entry on the challenges of scaling interpretability for complex AI models and methods for measuring the faithfulness of LLM explanations.

A diary entry introducing AI interpretability and discussing a paper on the limitations of sparse autoencoders for finding canonical units of analysis in LLMs.

A diary entry on Chapter 5 of the AI Safety Atlas, focusing on evaluation methods for assessing the safety and alignment of advanced AI systems, including benchmarks and robustness testing.