Discord AI moderation bug banned 8,000 users over harmless images
The system sees squares.
Hammering down. Image by Cybernews.
- Discord said a moderation bug wrongly banned more than 8,000 users over harmless image uploads.
- The bug caused immediate bans before human review, affecting spreadsheets, chessboards, game textures, and grid-like images.
- Discord says it fixed the issue and is restoring affected accounts, including about 200 banned over the weekend.
- The incident renewed criticism of AI moderation, especially as wrongful bans can disrupt users’ work and communication.
Key Takeaways by nexos.ai, reviewed by Cybernews staff.
Discord has admitted that more than 8,000 users were wrongly banned over the last two months. Harmless uploads, including spreadsheets, chessboards, game textures, and white or gray transparent images, were falsely flagged.
Around 200 more users were banned over the weekend before engineers identified the bug, which Discord says it has now fixed and is restoring affected accounts.
The platform, which is especially popular with gamers, said its moderation system compares uploads against databases of known harmful material before human review. A software bug caused accounts to be banned immediately, rather than waiting for the review.
"We're working on better safeguards so this can't happen again," Discord said in a post on X.
Our systems flag content by matching it against known harmful material. This kind of similarity matching can produce false positives, which is why a member of our Trust & Safety team always reviews flagged content before any action is taken.
undefined Discord Support (@discord_support) July 7, 2026
The intended behavior is to…
Why did the AI think the harmless images were dangerous?
AI moderation relies on pattern matching rather than understanding images as humans do, with Reddit and X users noticing many banned images of shared square grid patterns.
Gamers on social media speculated that the grid-like images could have potentially triggered AI into thinking that a poster may have something to hide, with grid-like layouts often deployed in an attempt to bypass AI filters.
One X poster recommended not sending any images until the vulnerability was fixed, adding that “AI moderation needs to be scrapped and replaced with humans.”
“This is exactly why companies shouldn’t be using clankers to decide whether accounts should be banned or not,”they continued.
Wrongful bans can affect not only casual gamers but also industry professionals. One game director expressed the inconvenience of being banned, explaining that it extends to all his communication for his job.
My account was wrongfully banned from your platform due to a bug in your AI automod detecting my GAME TEXTURES as CSAM. I need my account back as I'm a game director and use Discord for all my communication. I have requested a review of my suspension.@discord @discord_support pic.twitter.com/QfAkCIJo6S
undefined JDBRYANT 🎂 TODAY (@jdbryantdev) July 4, 2026
On June 30th, Discord also made waves by announcing that it is introducing selfies and government ID checks for age verification, despite privacy advocates citing concerns about biometric techniques.
Despite criticism, AI does get moderation right sometimes, as shown in a case where the TikTok AI moderator managed to flag an antisemitic post, even if human moderators overturned the initial flagging.
