A bug on X, previously Twitter, was inflicting quite a few posts over the weekend to be flagged as “Delicate Media,” thwarting the corporate’s personal makes an attempt to make its platform extra approachable to advertisers. In accordance with the X Security account in a submit, the problem has now been fastened and the group is engaged on eradicating the labels from impacted posts.
“Delicate media” is a label X makes use of to indicate content material that others might not want to see, like violence or nudity. X asks its customers who wish to recurrently submit such objects, to regulate their media settings to appropriately mark their photographs. As well as, there’s an possibility so as to add a one-time delicate content material warning to pictures and movies throughout X on iOS, Android, and the online. Posts may be marked as nudity, violence, or simply “delicate” to limit viewing behind a blurred content material warning that requires a further click on or faucet to view the media.
Nevertheless, X customers had been discovering that even innocuous pictures and media had been being marked as “delicate” — one thing the corporate itself can do, if it critiques reported objects after which chooses so as to add the label to the media to guard customers. Within the days earlier than Twitter’s acquisition by Elon Musk, a mixture of automation and human evaluate (members of the corporate’s belief and security group) would act on these reviews to make the choice.
On this most up-to-date occasion, the problem might have been the work of a spam bot, in keeping with a submit by Musk, which considerably contradicts the X Security announcement.
On Sunday, Musk wrote “An X spam/rip-off bot by chance flagged many legit accounts in the present day. That is being fastened.” An hour later, he reposted the message from the X security group which referred to the problem as a bug. The Security group posted afterward Sunday night time that “all impacted posts have been fastened and incorrectly utilized labels eliminated.”
The difficulty is the most recent misstep at X because it makes an attempt to determine a brand new monetization technique after advertisers fled the service beneath Musk. Talking at an occasion in November, X proprietor Elon Musk instructed advertisers to “go f*ck your self,” when questioned about massive manufacturers’ resolution to pause promoting on X over issues about antisemitic content material on the platform. The comapny was later mentioned to be pursuing small-to-medium-sized advertisers within the meantime because it strikes ahead with its technique to roll out AI and peer-to-peer funds in 2024.
The difficulty might even have been worsened by X’s workforce reductions, which impacted its belief and security group — the group that sometimes reviewed accounts for spam and delicate content material.
Bots flagging accounts isn’t the one space the place X has confronted points with elevated spam in current months. A seek for the phrase, “I’m sorry, I can not present a response because it goes towards OpenAI’s use case coverage,” on X lately revealed many automated accounts had been masked as common customers and, in truth, lots of them had been paying X Premium customers. Musk had believed that implementing a small charge would assist rid the platform of spam, however this indicated that a minimum of some bots had been keen to pay to look human. The corporate additionally admitted final summer season it had a Verified spammer downside when it introduced new DM settings that will transfer messages from Verified customers out of your inbox — one other indication that X’s Verification system was not hunting down spammers, as hoped.
X has skilled different main bugs since Musk’s takeover, together with a glitch that broke native photographs and hyperlinks in August 2023 and a international outage final month.