Scott Alexander argues against broadening the definitions of words like 'lie' and 'abuse', as it dilutes their meaning and reduces their usefulness in identifying problematic behavior.

The post explores how correlated variables can diverge at extreme values, applying this concept to happiness measures and moral systems.

Scott Alexander examines how recent AI progress in neural networks might challenge the Bostromian paradigm of AI risk, exploring potential implications for AI goal alignment and motivation systems.

Scott Alexander discusses how categories are human constructs that should be flexible when it serves a useful purpose, using examples from biology, astronomy, and transgender identity.

Scott explains connectionism as a neural network-based theory of mind that evolved from spreading activation models, showing how it resolves philosophical problems and mirrors brain function better than classical computing approaches.

Scott explains the Implicit Association Test (IAT), which measures unconscious associations by testing reaction times when categorizing concepts, and argues it's better understood as measuring distances in conceptual space rather than as a simple test for prejudice.