WhatsApp is free, so the millions of users on chatboards across the Internet must surely realise that on a service Meta is providing for free that they have every right to dump Meta AI on WhatsApp and you just have to suck it up and realise that it’s there to stay.
Meta AI on WhatsApp rolled out first in the US, but now it’s popping up in the UK, Europe and around the world, which is why there’s suddenly a huge interest in what it’s doing there, what it can see and frankly how much of your personal information might be shared with the AI.
And there’s no point trying to disable it, having searched the internet I did the only sensible thing and resorted to asking Meta AI on WhatsApp, how it can delete itself. It came back with some suggestions, none of which work. The button is going to stay hovering on your home screen and all you can do is to mute notifications and delete the Meta AI chat thread.
You might think that Meta AI is ok to have hanging around in the background, as we’re assured it can’t read your chats with your friends. It’s simply designed as a helper for your everyday life and Meta want to make sure that you can’t avoid finding their AI button before you think of using an alternative service like Gemini, Copilot or ChatGPT.
The thing is though, while Meta AI can’t read your WhatsApp messages unless you forward one to it, your end to end encryption now relies on recipients of your messages not sharing them with Meta AI as well. Oops, now it’s starting to sound quite sinister.
Meta simply say “Don’t share information, including sensitive topics, about others or yourself that you don’t want the AI to retain and use. Meta shares information with select partners so Meta AI can offer relevant responses.“
What Meta are really saying is don’t share anything with your WhatsApp contacts that you don’t want the risk that it might be shared around the world and used to target you with what they’ll call ‘relevant advertising’. Just because you don’t share your chats with Meta AI doesn’t mean that recipients of your messages might not.
I’m starting to see why some users are so worked up about AI being introduced to WhatsApp, which with it’s end to end encryption has seemed a pretty safe place to communicate. But it’s becoming very obvious why civilian chat services shouldn’t be used by the government and military – even if you don’t accidentally add a journalist to your chat, you don’t want the risk of state secrets being shared with an AI!
