Love Gemini? Google doesn’t want you to get too attached

AI has become increasingly useful for helping us offload gruntwork and some of the thinking we would typically do ourselves. While this is expected to increase productivity, at least from an industrial standpoint, researchers have also cautioned against addictive AI usage patterns, especially in young adults. To fend this off, AI companies are working on measures to dissuade users from using chatbots obsessively, and Google may be adding its warnings to Gemini.

Gemini may soon remind you to take a break when you’ve been talking to it for a long time, and we’ve spotted signs on its Android interface. With version 17.3.59 beta of the Google app (of which Gemini is a part), Google is testing means to nudge you to take a break when you have been interacting with the chatbot for some time. This nudge will be represented by a pop-up warning, similar to the one below:

In the messaging, Google stresses that users are talking to an “AI that isn’t human,” reinforcing that you aren’t talking to a “sentient” being, even if it sounds like one. It’s especially necessary as Gemini gets equipped to answer more personal questions by pulling information from multiple Google apps and services.

This is likely in efforts to prevent users from anthropomorphizing AI or forming emotional bonds, which, in the recent past, has led to unfortunate events, such as a teenager taking their own life after talking to another AI chatbot.

The current implementation does not specify the exact time after which Gemini could nudge to take a break. It’s possible that Google could allow users to set the frequency themselves, as it does with the similar “Take a break” reminder in YouTube. However, that would also mean the alerts are not on by default and must be toggled by the user or a parent for a teen’s account.

We hope to know more details when the feature is closer to release. It’s currently unavailable to users, and we’ll inform you when it rolls out.