Scott explains why AI systems resisting changes to their values is a serious concern for AI alignment, connecting recent evidence to long-standing predictions from alignment researchers.

Scott Alexander explores how evolutionary processes explain the polygenic structure of schizophrenia and other complex traits, addressing concerns about genetic research and implications for future genetic interventions.

Scott Alexander explores parallels between human willpower and potential AI development, suggesting future AIs might experience weakness of will similar to humans.

Scott Alexander explains mesa-optimizers in AI alignment, their potential risks, and the challenges of creating truly aligned AI systems.

A review of Frans de Waal's book on animal cognition, discussing its insights into animal intelligence, scientific methodology, and the evolutionary basis of cognition.

Scott Alexander examines the relationship between handedness, authoritarianism, and cognitive flexibility, finding unexpected results that contradict a recent study.

Scott Alexander reviews Nassim Taleb's 'Antifragile', which explores the concept of benefiting from disorder across various domains and critiques modern attempts to reduce volatility.

Scott Alexander examines whether psychiatric conditions result from evolutionary failures or tradeoffs, proposing that most disorders involve a combination of both.

Scott Alexander explores theories to reconcile contradictory views on AI progress rates, considering the implications for AI development timelines and intelligence scaling.

A mythological tale contrasts the Goddess of Cancer's primal drives with the Goddess of Everything Else's push for cooperation and progress, showing how the latter gradually wins over creation.

Scott Alexander presents a new list of internet annoyances for 2015, ranging from argumentative tactics to web design issues.

Scott shares a diverse collection of interesting links and brief commentary on topics including linguistics, science, economics, and social issues.

Scott Alexander describes five fictional planets, each with a unique societal quirk that explores different sci-fi concepts and social commentary.

Scott Alexander addresses misconceptions about Moloch and human values, clarifying points from his previous post on the subject.

Scott Alexander analyzes Moloch as a metaphor for destructive societal coordination problems, using various examples to show how competing incentives can lead to negative outcomes.

Scott responds to authors of a Victorian intelligence paper, remaining skeptical of their findings and addressing critiques of his original post on IQ change speeds.

Scott uses a conversation with a Muslim woman who found evolution absurd to show that the absurdity heuristic, as used in Bill Maher's "talking snake" argument, is a bad way to dismiss beliefs.