An important use-case is federated learning, which Google, and many healthcare / pharmaceutical companies are very interested in. In federated learning, multip

An important use-case is federated learning, which Google, and many healthcare / pharmaceutical companies are very interested in. In federated learning, multiple companies or groups with their own private data come together to train a model jointly on all the private data, while keeping the data private. You need more than zero-knowledge proofs to actually do federated learning securely, but to my limited knowledge they are one tool in the toolbox that can be useful.