If you've been on the Facebook app lately, you've seen Meta's AI inject itself into the comments section with summaries of what people are saying. Given how wild Facebook comment sections often get, it's not hard to imagine how funny some of these summaries turn out to be. (This isn't the first time the meta's AI has appeared in the comments section, by the way: 404 media Saw him pretending to be a parent in a Facebook group.)
After seeing screenshots of the feature shared on threads and Reddit, I decided to check out the comment sections on my Facebook app. I found AI summaries popping up on many of the posts I checked — unsolicited answers and all. An AI abstract on a post about the store closings states, “Some commenters call the store closings a 'woke' or poor choice, while others point to the rise of online shopping.”
Another Facebook post deputy A comment section about Mexican street wrestlers summarized that some people were “less than impressed” with the performance and called it “a stupid way to fight back”. AI also picked up some light-hearted jokes from people about bobcat sightings in a Florida town. “Some people applauded the sighting, with one commenter hoping the bobcat remembered the sunscreen.”
It's not yet clear how Meta chooses which posts to display comment summaries, and the company did not immediately respond. the edgeRequest for comment.
Either way, the summaries don't really add anything that I find useful (unless you like vague notions of what random people say) — but it gives you a sense of these posts. Can help identify where the comment section has become too toxic to bother scrolling.
AI summaries have also raised privacy concerns, as Meta is feeding user comments into its AI system. Over the past week or so, many Facebook and Instagram users in the EU and UK received a notification informing them that Meta would be training its AI on their content. (Data protection laws in both regions require Meta to disclose this information.) Although Meta will let users object to their data being used to train AI, the process is relatively simple. No, and the company has rejected some customer requests.
Here in the US, Meta's privacy policy page says the company uses “information shared across Meta's products and services,” including posts, photos and captions, to train the AI. Meta allows you to submit a request to correct or delete personal information used to train its AI models, but this only applies to third-party information. Everything else seems fair game.