Google’s Gemini AI has started interacting with your third-party apps, opening the door for the search company to “[f]ind relevant information” and “perform actions for you” based on those apps. Of course, Google hasn’t made it an opt-in thing.
There are reasons for this, the most likely being that you can be claimed as an active Gemini user simply by using your device as you normally would. That looks good on monthly user figures, which helps justify all the AI spending the company is doing.
Just like a Gemini
The company recently shared an email explaining what the new integration is all about. Part of the explanation includes details about human reviews, who “(including service providers) read, annotate, and process your Gemini Apps conversations”. When those conversations include details of “Phone, Messages, WhatsApp and Utilities”, it may be time to get worried.
There are ways to prevent Gemini from peeping on what you’re doing on non-Google services (and even Google ones, but that will limit functionality and perhaps kill it entirely). Taking a trip to this page when signed into your Google account will let you know exactly which apps are sharing data with Google Gemini and, if needed, turn that sharing off.
Even then, Gemini will remain in place and continue to monitor what you do during interactions with Google. The company retains those interactions on a specific page. While they may be turned off and the current history deleted, “chats are saved in your account for up to 72 hours, whether Gemini Apps Activity is on or off.”
Ours was set to retain chats for eighteen months by default, and a glance back through the log shows how little users have to do to be considered a Gemini user. Intermittent instances of “Used an Assistant feature”, with no further explanation, and several mistaken prompts (as in, Assistant accidentally overheard conversations) were logged as interactions. The latter included the exact wording that was overheard as a “Prompt”.
