The Fog of Trust protocol was supposed to allow a prover Peggy and a verifier Victor to perform a secure multiparty computation to determine the number of third parties that Victor trusted and that vouched for a certain property of Peggy. We intended to use formal methods to first state and then prove the privacy properties of the protocol. Instead, the analysis lead to the discovery of two previously unknown design failures that allow an adversary to break the privacy assurances the protocol was expected to provide. This paper briefly presents the Fog of Trust protocol and the vulnerabilities, which at this point we are unable to fix.