Reference/sample instance for the Continuous Research framework

Interactive visualization of the incentive cap model — why AI companies underinvest in safety relative to societal optimum

A substrate for treating research as a living, version-controlled, mergeable artifact — autonomous agents propose updates as new data, literature, and reviewer comments arrive, and a managing author accepts them through a Git-native review workflow. This is a framework; individual research efforts are instances built on it.

Forecasting AI job disruption and escalation by applying a 3-part economic demand framework (Bounded, Unbounded, Adversarial) to Anthropic and O*NET employment data.