Featured Project · AISC 2026

Doom Assumptions

An interactive map of AI risk arguments. Set your own probabilities, compare named worldviews (Yudkowsky-like, LeCun-like, Hinton-like, …), and find cruxes automatically.

Open the tool →

Articles

Some AI threats people aren't thinking about AI identity is not tied to its model Stockfish is not a chess superintelligence (and it doesn't need to be) Estimates of the expected utility gain of AI Safety Research The quest for general intelligence is hitting a wall [April Fool's] AGI both does and doesn't have an infinite time horizon AI loves octopuses Ranking the endgames of AI development LLMs are stuck in Plato's cave Agents, Simulators and Interpretability