Twitter nsfw/hate accounts in feed

Septimiu

Hi,

On my feed in Twitter, I get sometimes NSFW content from fake channels that doesn't go away.

I noticed there might be cross-origin loading of posts or at least the media.

I go into more detail in this post on the Avast community, but got no answers…

Is the Web Application Firewall of Bitdefender endpoint also not able to prevent these sources? I know it's a free speech platform and this isn't against the rules, but I always "Not interested" and block, and still get them… is it tied to my account only because of some compromise? Note that I do have 2FA enabled.

Help is appreciated

Alexandru_BD

Hello,

A typical antivirus firewall or security suite generally does not block legitimate traffic from X’s servers, so it won’t stop NSFW media from appearing in your feed. The firewall doesn't have post categories settings to block in this scenario. I think this type of issue is usually not a sign of a compromised account. It’s more likely tied to X's recommendation system, accidental interactions (clicks/likes/follows), or new/duplicate spam accounts that bypass your blocks.
Since you are actively choosing to visit Twitter or use their mobile app, the firewall sees that as legitimate behavior. It usually allows all images and videos from those domains. Also, NSFW media in your feed is not (by itself) a virus or malware distribution. So, the antivirus has no real reason to block it on the basis of cybersecurity threats.

Septimiu

I see, I think you are right in that if the company rightfully decides to have multiple origins for different content types, that is their responsibility and oversight, and the AV cannot notify them of infringement as long as minimal security best practices are enforced. That being said, if the AV did not detect any configuration issues, I trust that they don't exist.

Thank you very much.

Edit: to clarify, blocking the Fastly domain blocked only nsfw content, but legitimate videos from trusted accounts still worked.

Alexandru_BD

Yes, if I remember correctly, the platform used to have a “muted words” feature (or mute settings). Perhaps you can add certain keywords that typically appear in NSFW or spam tweets. This in theory can automatically hide content containing those terms. And keep clicking "Not interested" can help refine your feed, theoretically. Another thing to check would be that there aren’t any suspicious or inadvertently followed accounts. Even if you think you never followed them, a compromised session or a misclick can add unwanted accounts.
I think the algorithm can sometimes ignore user feedback or could be slower to adapt. Also, if any of your followed accounts are liking or retweeting NSFW content, X may still show it to you.
When you block or report an account, the spammers often create new handles or fake channels to bypass blocks. You end up seeing very similar content from what appear to be different senders.
Furthermore, if you accidentally clicked on NSFW tweets (even to read replies), X’s algorithm may interpret this as interest.

I hope this helps.

Regards