Late 2021 MIRI Conversations

Embedded Agency

AGI safety from first principles

Iterated Amplification

Value Learning

Risks from Learned Optimization

Cartesian Frames

Intro to Brain-Like-AGI Safety

Agency: What it is and why it matters

Thoughts on Corrigibility

Epistemic Cookbook for Alignment

AI Safety Subprojects

Modeling Transformative AI Risk (MTAIR)

Practical Guide to Anthropics

The Causes of Power-seeking and Instrumental Convergence

Finite Factored Sets

Anthropic Decision Theory

Reviews for the Alignment Forum

Predictions & Self-awareness