Measuring AI Risk with Cost‑Effectiveness
By assigning a unit of “goodness” to incremental improvements, the framework enables donors and policymakers to compare diverse projects on a common scale, fostering more strategic allocation of global resources toward long‑term safety.