Facebook, Instagram opt out of allowing Apple AI to scrape their data for training

Apple Intelligence is upon us, but not every company that owns a major website is a fan of Apple scraping its data to train AI.

A new report by Wired found that several widely recognizable online brands have chosen to opt out of training Apple’s AI bot. Among those who have deliberately turned it off are Facebook, Instagram, the New York Times, Craigslist, and Condé Nast. If the opt-out request is truly honored, Apple Intelligence cannot and will not get any data from those sites until, or unless, any of them enter into official business agreements with Apple for the use of their data.

Wired‘s report goes into the nitty gritty technical details of how these data scraping bots work, but the main thing to know is that it’s similar to how Apple has trained things like Siri over the years, but now with an extra layer of possible copyright infringement that has made companies wary. Apple, to its credit, started offering a tool to opt out of data scraping months ago, so it’s not like these companies had to figure out a way to do it themselves.

Considering so many major companies are turned off by the proposition of feeding data to Apple’s AI, and the fact that Apple needs to give its AI literal pep talks to stop it from making things up, one could reasonably wonder if maybe we jumped into this lake a little too early.