LW1.0 username Manfred. Day job is condensed matter physics, hobby is thinking I know how to assign anthropic probabilities.
Humans aren't agents - what then for value learning?
How to get value learning and reference wrongΩ
Philosophy as low-energy approximation
Can few-shot learning teach AI right from wrong?
Boltzmann Brains and Within-model vs. Between-models Probability
[Link]Is this what FAI outreach success looks like?
Book Review: Consciousness Explained
A useful level distinction
Explanations: Ignorance vs. Confusion
Empirical philosophy and inversions