Can someone explain what this actually means? I assume this still runs on Google's cloud so it's not 'private' in any meaningful sense.
Differentially private means that:
training_algorithm(training data with a row that has "ForHackernews blood test report...") hard to distinguish from training_algorithm(training data without that) upto a factor of epsilon. They have explained further in the article itself with concrete values for epsilon.
It does not run on Google’s cloud. You can download the model and host it yourself, locally or using a provider you trust.