"Since the AI model could potentially be reading sensitive information or documents, Google says it is addressing privacy concerns by stating that content from Gmail, Docs, and Drive will not be viewed by human reviewers, and it will not be used to show ads or train future versions of the Bard AI model."
"But Bard’s chat also acknowledges that it may ask to analyze your messages 'to understand the context of your conversations, your tone, and your interests.' . . . And it may 'analyze your message history with different contacts to understand your relationship dynamics' . . ."
This means that this is non-consensually reading unwitting correspondents' messages, as well.
AI reviewers can be just as oppressive and biased as human ones. AI isn't very astute with "tone" nor with idiomatic expressions. (I got locked out of two Twitter accounts because of that.)
Now I have to not electronically communicate with anyone using an Android phone nor with Instagram installed on their device. I also have to get them delete all my previous messages and e-mails to them.