The limits of counterfactual computation

Abstract

We show that the protocol recently proposed by Hosten et al. does not allow all possible results of a computation to be obtained counterfactually, as was claimed. It only gives a counterfactual outcome for one of the computer outputs. However, we confirm the observation that the protocol gives some protection against decoherence. In some situations, though, it may be more effective simply to run the computer several times.

Hosten et al. have responded to the second version of this paper, and the present version includes our reply to them. The text is otherwise unchanged