'Goodbye Meta AI' is the next major spam chain mail, and no it doesn't stop Facebook scraping your data
"Share this with five friends before midnight or you may die" is a sentence that runs down the spine of many an ex-MySpace user. Not because they narrowly avoided death but because of the cringey feeling of believing that would ever work. Seemingly to counter Meta training its AI model on public posts, many are falling for chain mail once more.
A new post going around Facebook and Instagram that starts with "Goodbye Meta AI" claims that those who post this and the message that follows are essentially opting out of Meta using their data to train its AI model (via BBC). This is, of course, untrue, and speaking to the BBC, a Meta spokesperson said "Sharing this story does not count as a valid form of objection".
The message continues: "Please note an attorney has advised us to put this on, failure to do so may result in legal consequences. As Meta is now a public entity all members must post a similar statement". Yeah, that's spam chain mail if ever I've seen it.
If you want to opt out of Facebook using your data to train its AI, you have to go into the app or site itself, open up "Settings and Privacy", then click "How to manage your information". When asked if this is in relation to AI, you are prompted to give a reason as to why Facebook using that data affects you. A helpful stranger on Reddit suggested writing the following:
"It violates my right to privacy as provided by the European Convention of Human Rights. My data was originally provided for specific purposes and changing that purpose goes against GDPR."
What is artificial general intelligence?: We dive into the lingo of AI and what the terms actually mean.
When I filled this out, Facebook confirmed it would honour my objection, though it's unclear if this paragraph has anything to do with it. Given filing your response is technically optional, it could just be a way for Meta to understand users' objections.
The fact that you have to read through Facebook's rationale for its AI and then potentially reason your case seems absurd to me, when you can opt in and out of many uses of your data, and change how advertisements work in just a button press.
Given many are falling for this chain mail—and Meta specifically asks if objections are in relation to AI—it all suggests that AI being trained on posts isn't a hugely popular prospect. Luckily, opting out isn't too hard once you know what to do.