Scott examines a paradox where bloomers and anti-doomers warn against apocalyptic thinking while treating doomerism itself as an unprecedented existential crisis requiring drastic action. Longer summary

Scott analyzes Jason Pargin's novel 'I'm Starting To Worry About This Black Box Of Doom' and similar arguments from Peter Thiel, Tyler Cowen, and progress studies advocates, noting they all share a contradiction: they argue we should stop treating problems as unprecedented crises, except for the problem of treating problems as crises, which they present as uniquely dangerous. The post examines how these thinkers simultaneously advocate against doomerism while expressing extreme concern about doomerism itself. Scott argues this reveals that strong views about a 'crisis of doomerism' are incompatible with worldviews that reject the existence of real crises, and concludes that optimism should be a heuristic rather than an absolute principle, with problems (including excessive doomerism) evaluated based on evidence using consistent standards. Shorter summary

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 responds to Tyler Cowen about USAID funding, correcting his own previous claims about overhead costs while maintaining that Cowen's criticism of USAID was misleading and potentially harmful. Longer summary

Scott Alexander responds to Tyler Cowen's criticism of his previous post about USAID funding. He addresses several points: whether Cowen endorsed Rubio's claims about USAID waste, the true nature of overhead costs in USAID-funded organizations, and the broader debate about foreign aid effectiveness. Scott shows that actual administrative overhead in major USAID partners like Catholic Relief Services is much lower than previously thought (around 6-7% rather than 30%), admits his mistake on this point, but maintains his criticism of Cowen's original post as misleading. He argues that USAID's work is predominantly focused on essential humanitarian aid rather than wasteful programs. Shorter summary

Scott Alexander criticizes Tyler Cowen and others for misrepresenting USAID's funding model, explaining how regranting through other charities is both necessary and effective despite seeming inefficient to outsiders. Longer summary

Scott Alexander criticizes a Marginal Revolution post by Tyler Cowen about USAID funding, where Cowen suggests that only 12% of funds go to recipients. Scott explains that this is misleading because USAID is not a direct charity but a funding organization that works through other charities. He details how the grant-making process works, defends the overhead costs, and points out that Cowen himself runs an organization (Mercatus Center) that does similar regranting. Scott particularly criticizes Trump and Rubio for misrepresenting these programs as wasteful, noting that programs like PEPFAR have saved millions of lives and have very low rates of unexplained expenses. Shorter summary

Scott Alexander critiques Tyler Cowen's use of the 'Safe Uncertainty Fallacy' in discussing AI risk, arguing that uncertainty doesn't justify complacency. Longer summary

Scott Alexander critiques Tyler Cowen's use of the 'Safe Uncertainty Fallacy' in relation to AI risk. This fallacy argues that because a situation is completely uncertain, it will be fine. Scott explains why this reasoning is flawed, using examples like the printing press and alien starships to illustrate his points. He argues that even in uncertain situations, we need to make best guesses and not default to assuming everything will be fine. Scott criticizes Cowen's lack of specific probability estimates and argues that claiming total uncertainty is intellectually dishonest. The post ends with a satirical twist on Cowen's conclusion about society being designed to 'take the plunge' with new technologies. Shorter summary