Resources
AI Safety
There already exist tons of information and resources on AI safety on the internet. To avoid reinventing the wheel, I recommend starting with these two lists: Victoria Krakovna’s AI safety resources and Sudhanshu Kasewa’s AI safety cheap tests and resource lists. Below, I’ve added some resources in areas they don’t cover as much.
Doing research well
- Tips for Empirical Alignment Research Ethan Perez
- Research as a Stochastic Decision Process Jacob Steinhardt
- Tips and Code for Empirical Research Workflows John Hughes & Ethan Perez
- How I Think About My Research Process Neel Nanda
- Highly Opinionated Advice on How to Write ML Papers Neel Nanda
- How to Become a Mechanistic Interpretability Researcher Neel Nanda
- ML Safety Research Advice Gabriel Mukobi
- AGI Safety Career Advice Richard Ngo
- Some Advice on Independent Research Marius Hobbhahn
- A Model of Research Skill
- Practicing AI Research Jason Wei
- A Woefully Incomplete Guide to Technical Upskilling
- Cargo Cult Science Richard Feynman
Beyond technical research
Grantmaking
Strategy
- AI Strategy and Policy Career Review 80,000 Hours
- Forethought research A research group working on macrostrategy for the transition to advanced AI.
- A Case for Strategy Research Rozendal, Shovelain & Kristoffersson
- Important, Actionable Research Questions for the Most Important Century Holden Karnofsky
Field building
Staying current
Newsletters
Life Advice
- College Advice for People Like Me Henry Josephson
- How to Not Forget What Matters Henrik Karlsson
- What’s Stopping You? Neel Nanda
- How I Practice at What I Do Tyler Cowen
- Useful Vices for Wicked Problems Holden Karnofsky
- Tempering Hardcoreness Ajeya Cotra
- Love is a skill, not a feeling Alain de Botton