Hidden motives or innocent failure? wp.me/pAgEM-16T

Improving the territory takes a good map, a part of the territory particularly worth improving and depicting in detail.

Thinking involves updating your map on the state of your map.

The more you care about, the bigger your world becomes.

The brain is an organ that senses mathematics.

Epistemic rationality is about forming accurate understanding of the decision problem, not just the world.

"Because" conflates explanation and justification. Use with caution.

You are not your mind; you are the one who comes after, observing, and one who comes before, determining.

Compartmentalization is sustained by cached hidden inferences of propriety of context.

Decisions are self-fulfilling beliefs; actions are self-fulfilling evidence.

Nick Bostrom's Fable of the Dragon Tyrant in Philosophy Now: philosophynow.org/issues/89/The_…

A game composes players' responses to their beliefs with the dependence of their beliefs on each other's responses.

Concepts are inference applicability heuristics; ruling X an instance of Y without expecting novel inferences is cargo cult classification.

A consequentialist formal system decides which theorems to have based on the hypothetical consequences of their truth about its models.

An agent controls its output without controlling its program. Could the value of the world be controlled without control over its state?

When asked to be specific, give a description that abounds with implicit non-trivial consequences.

Human value depends on the world. Running a theorem prover establishes a dependence of human value on the theorem prover's theory.

Why is the word "dreams" used to describe both pseudorandom nocturnal hallucinations and also heartfelt aspirations for real life?

Worth reading Superintelligence by Bostrom. We need to be super careful with AI. Potentially more dangerous than nukes.

"Safe" is a wrong word for describing a process of rewriting the universe.