Scott explores three approaches to 'writing for AI' - teaching knowledge, influencing beliefs, and enabling simulation - finding the first limited, the second theoretically confused, and the third creepy and ethically troubling. Longer summary

Scott examines the concept of 'writing for AI' - creating content that will influence future AI systems - through three lenses: helping AIs learn knowledge, presenting arguments to shape AI beliefs, and helping AIs model writers in enough detail to recreate them. He finds the first two either limited or theoretically muddled, and the third deeply unsettling. The post explores why influencing AI beliefs faces both practical obstacles (alignment training will override corpus data) and theoretical ones (finding the right sweet spot of influence). Scott is particularly disturbed by the idea of AIs simulating him, comparing it to being 'an ape in some transhuman zoo,' and struggles with questions about whether writers should try to impose their values on future AI systems. Shorter summary

Scott Alexander shares 35 diverse links covering scientific studies, historical facts, cultural observations, and technological developments. Longer summary

Scott Alexander presents a collection of 35 interesting links on various topics. These include scientific studies, historical anecdotes, cultural observations, and technological developments. Some notable items are: a failed replication of the link between childhood screen time and attention problems, insights into the Canadian tech scene, a map of Reddit, and a technology for animating old photos. The post also covers topics like urban planning, economic policies, and quirky historical facts. Shorter summary