There’s an irony to the backlash that embroiled Meta when it added its infamous blue AI circle to WhatsApp. It seemed Meta AI was a mandatory guest lurking in your chats and could not be removed. It can be blocked from your chats, though, details here.
Meanwhile, both Google and Microsoft have demonstrated the high-risk in adding platform-wide AI to our devices. Suddenly, the content we consider private and secure is anything but, AI has a thirst to see all we’re doing and we can’t stop it.
Microsoft went first, with its fabled Recall upgrade to Windows 11. As has been widely reported, this sci-fi photographic memory that snapshots everything that’s done on a PC and stores it within a searchable local repository will read any text messaging windows open on screen. Only Signal has managed to stop this.
Then Google went next. Gemini, it announced, would start accessing third-party apps on phones, including Messages and WhatsApp. Initially it seemed that the data vacuumed up in its midst would even be used for AI training, but that’s not the case.
Google told me the update “is good for users,” as Android phones can now “use Gemini to complete daily tasks on their mobile devices like send messages, initiate phone calls, and set timers while Gemini Apps Activity is turned off.”
However, as Neowin warns, while Google promises that under normal circumstances, Gemini cannot read or summarize your WhatsApp messages. But, and this is a big but, with the ‘help’ of the Google Assistant or the Utilities app, it may view your messages (including images), read and respond to your WhatsApp notifications, and more.”
Practically, you need to tell Gemini to access WhatsApp by prompting the AI assistant to read or reply. But technically, the AI engine is not blocked from accessing the app. Not quite as bad as Recall, but we’re entering a new world of blurry privacy lines.
Google trumpeted its update to enable Gemini to access third-party apps without using data for training, but as Ars Technica warns: “No, Google, it’s not good news.” The issue, confirmed by Neowin, is that there doesn’t seem to be a way to disable this.
“Google marketers may claim the integration is good news,” Ars Technica says. “A significant number of [users] don’t want Gemini or other AI engines anywhere near their devices. For the time being, these users are being left completely in the dark.”
You can restrict Gemini within your app settings, but the onus is on you to fiddle with your phone settings to opt out. That’s not how this should be done.
Whether Windows or Android, you should act now and at the very least check your settings and be aware of what content is being read, stored and used for training on your device. Also bear in mind that with messaging, this AI is also reading and possibly storing or even human reviewing other user’s content — often without them realizing.
