If you’re actively using Meta’s Threads app, you’ve probably encountered some level of engagement bait in the “for you” feed. Now, Instagram chief executive Adam Mosseri has acknowledged the issue and said the company is “working to bring it under control” after a surge in such posts.
“We are seeing an increase in engagement baiting on threads and are working to control it,” Mosseri said in a post on the thread. He did not say what action the company was taking, but said “further action will be taken” regarding the issue.
Mosseri’s comments are the first time a Meta executive has acknowledged the issue, which has become an increasingly prominent topic of discussion on the thread. The app defaults to an algorithmic “for you” feed, so even if it’s the type of content that many users say they don’t want to see, engagement-bait style posts that get a lot of responses are not allowed on the app. It is often spread.
However, this issue can be difficult to address because what people often refer to as “engagement bait” takes many different forms within apps. There are some downright spam posts (here’s a particularly egregious recent example) lifting up the kind of copy-paste content that’s been widely shared on Facebook. I have an account that posts free-form AskReddit style questions. And some posts are outrageous, as Business Insider’s Katie Notopoulos recently reported.
These posts often touch on polarizing topics, with the sole purpose of eliciting angry reactions from other users who have had controversial opinions from someone they don’t know shoved into their timelines. I think so. For example, I’ve seen dozens of posts and hundreds of replies about whether young children should be allowed on planes. In Notopoulos’ experiment, a post about not feeding visiting children (heavily inspired by a viral tweet and Reddit post from two years ago) received more than 1 million views and 5,000 comments. Got a reply.
Mosseri and Meta also didn’t explain why they found it easy to leverage Threads’ algorithms to spread the word on this type of content, but it’s because the app shows users This seems to be related to how the responses are prioritized when determining the response. “Not all comments and replies are good,” Mosseri said.
