Over the weekend, X, formerly known as Twitter, encountered a technical glitch that led to a significant number of innocent and clear posts being inaccurately labeled as “Sensitive Media.” This small problem posed challenges for users and disrupted the platform’s efforts to create a more advertiser-friendly environment.
The X Safety account promptly acknowledged the issue, assuring users that the bug has been resolved and the team is actively working to remove the erroneous labels from affected posts.
“Sensitive Media” is a designation used by X to flag content containing potentially objectionable material such as violence or nudity.
X wants to be the safe space for everyone, hence the “Sensitive Media” tag for violence or nudity. Normally, you choose how to mark your own posts. The platform encourages users who frequently share such content to adjust their media settings accordingly but for reported content, X sometimes steps in.
Additionally, users have the option to include a one-time sensitive content warning on photos and videos across X’s platforms, requiring an extra click or tap to view the media.
Unexpectedly, users found that even harmless photos and media were being incorrectly marked as “sensitive.” Typically, this labeling is done by X after a review of reported items, with the company deciding whether to add the label to protect users.
Elon Musk, the figure behind X’s recent acquisition, suggested that a spam bot might be behind the issue, somewhat diverging from the initial explanation provided by the X Safety announcement.
In a tweet on Sunday, Musk acknowledged the situation, stating, “An X spam/scam bot accidentally flagged many legitimate accounts today. This is being fixed.”
Despite this, the X Safety team maintained that the problem was a bug and not the result of a spam bot. Later that night, the Safety team assured users that all affected posts had been rectified, with incorrectly applied labels removed.
This situation adds to X’s difficulties as it adjusts its way of making money after Elon Musk bought the company. In November, Musk’s reaction to advertisers asking about the platform’s content caused problems, and some advertisers stopped using X and suspending their services on X.
Now, X is concentrating more on smaller advertisers and getting ready to bring in AI and peer-to-peer payments in 2024.
The Reduction in X’s workforce, particularly in the trust and safety team responsible for reviewing accounts for spam and sensitive content, might have made the problem worse. This recent glitch underscores the ongoing challenge of finding the right balance between user experience, content moderation, and the platform’s financial objectives.
In Addition to the labeling problem, X has been dealing with a rise in spam lately. Even though Musk introduced a small fee to discourage spam, automated accounts were found pretending to be regular users, including among those who pay for X Premium.
Last summer, the platform encountered issues with its Verification system when it was discovered that spammers had infiltrated Verified accounts. This led to changes in Direct Message settings.