About Us Features Pricing Log in Start Free Trial

Automated Moderation Failures: Threads Banned Us

We previously discussed the growing risks of relying entirely on artificial intelligence for digital scanning and content moderation, particularly in the context of proposals like Chat Control. Automated systems are fundamentally prone to false positives, and yesterday, we experienced those consequences firsthand.

Our official Threads account was suspended shortly after we published a routine post highlighting custom interface options in MevaSearch:

Did you know you can fully customize the UI in MevaSearch completely visually? 🎨
No CSS, no coding, just toggle settings to tweak accent colors, custom backgrounds, fonts, button shapes, and themes in real time.
You can even export/import JSON settings or upload community themes! 🚀


No explanation or policy violation reason was provided. Instead, Meta required a phone number verification step. When we entered our number, the verification code failed to arrive. Shortly after, the platform blocked the number entirely for exceeding the maximum number of attempts—despite the fact that zero messages were ever delivered, even to spam folders.

Trying to resolve the issue exposed a systemic vulnerability: the complete absence of human oversight in customer support. Meta offers no telephone helpline and no live chat with human moderators. All support requests are handled exclusively by automated AI assistants that provided non-functional links and offered no meaningful assistance, leaving our profile locked with no recourse for an appeal.

This incident illustrates a broader challenge in modern digital infrastructure: as moderation and support are fully automated, administrative errors leave users with zero accountability or paths to resolution.