Meta reportedly working to restore falsely banned accounts, acknowledges excessive blocks in CSE crackdown

Meta is reportedly working to fix a widespread issue that has led to countless users having their accounts falsely banned. The tech giant has apparently acknowledged that an overzealous crackdown on child sexual exploitation (CSE) material led to an excessive number of blocks. This development comes after a wave of user complaints and growing pressure.

For weeks, users around the globe have been reporting sudden suspensions of their accounts, with Instagram (and sometimes Facebook) citing that they violated the platform’s guidelines and were accused of posting CSE material. We first highlighted this major concern on June 5. The bans seemed to hit without warning, leaving many scrambling to understand what went wrong.

The situation escalated quickly, with frustrated users even forming groups to collectively petition Meta and explore legal action. We covered this growing movement in a follow-up article that you can read here. It was clear that many users were affected, however, the lack of coverage from mainstream media outlets left users even more frustrated as many felt like they have no hope.

But it now seems that there’s finally some light at the end of the tunnel. A Korean National Assembly member, Choi Min-hee, who chairs the Science, ICT, Broadcasting, and Communications Committee, recently shared an update. They revealed that their office reached out to Meta Korea seeking answers. Meta Korea then confirmed that a global crackdown on child and youth pornography-related activities on their platforms led to some user accounts being “excessively blocked.” They stated they are aware of the issue and are in the process of sequentially restoring these accounts, while also trying to pinpoint the exact problems.

While the confirmation came through a Korean official, it’s suggested that the restoration efforts are likely global, given the nature of Meta’s worldwide crackdown. This news offers some reassurance to users who felt their accounts were unfairly targeted. It seems Meta has a lot of work ahead to undo the damage and regain user trust, but at least they’ve acknowledged the problem.

That said, this is yet another example of why AI content moderation has a long way to go. Hopefully, platforms are able to figure out a way to prevent such situations sooner rather than later.