> the VNM axioms assume that preferences aren't over histories at all, but instead over states That depends on how you formulate the axioms; or, equivalently,

> the VNM axioms assume that preferences aren't over histories at all, but instead over states That depends on how you formulate the axioms; or, equivalently, how you formulate the state. Eg, you can construct a vNM agent that strongly avoids "outcomes" where it spend a week awake, even if the end state of the world the agent is operating in retains no information about that. The "outcomes" A_i of the vNM axioms are (potentially) the entire observable history of the agent's universe.