Machines of Loving Grace
A vision essay on opportunity, responsibility, and advanced AI.
Read on AnthropicCurated Reading List
Constitutional AI
Foundational work on principle-guided training for safer assistant behavior.
Read on arXivAI and Compute
A frequently cited analysis of compute trends and AI progress.
Read on OpenAIConcrete Problems in AI Safety
Early practical framing of concrete research questions in AI safety.
Read on arXivLanguage Models (Mostly) Know What They Know
Research on calibration and uncertainty in language models.
Read on arXivToward Monosemanticity
Interpretability work focused on clearer internal model features.
Read articleFor broader discovery: Google Scholar search for Dario Amodei