Over the past eight months, we have been working on a new approach to nuclear warhead verification, generously funded by Global Zero. As part of this effort, we are seeking to develop and demonstrate the proof-of-concept for an inspection system that, by design, cannot divulge any classified information … because it is never measured. In cryptography, this is called a a zero-knowledge proof. In such a proof, Person A (the host) proves to Person B (the inspector) that a proposition is true without revealing why the proposition is true, i.e., in this case, that a genuine warhead is presented for verification or dismantlement without revealing any design information about it. One of our collaborators is Boaz Barak, a cryptography expert and Senior Researcher with Microsoft Research, New England. Earlier this year, Boaz has given talks focused on some of the conceptual ideas at the Institute for Advanced Study in Princeton (announcement and video) and at the Newton Institute in Cambridge, UK (video). The project will involve experiments and measurements that are currently being set up at the Princeton Plasma Physics Laboratory. We will regularly report on the progress.
Here are also the slides from an earlier but longer talk given at PPPL.