[Spam] Verbatim-duplicate Chinese spam replies evade detection at scale

[Holipori]

Summary

Chinese-language reply spam floods the replies of virtually every high-engagement Chinese-language post on X. The spam is trivially identifiable — many accounts posting the exact same sentence verbatim within hours — yet none of it appears to be caught by the safety pipeline (botmaker / scarecrow / abuse-enforcement-service in this repo). Reports on these accounts also appear to go nowhere.

Observed patterns (screenshots attached)

Pattern 1 — verbatim duplicate reply farms. The identical sentence 「比她好看的没她骚，比她骚的没她好看」 (and a first-person variant 「比我好看的没我骚…」) posted by many unrelated accounts under the same viral posts. Examples from a single reply thread within ~6h: @marrenz113689, @dignitymalik, @danna_var_her — mixed Chinese/Western display names, recycled profile photos, reply-only activity.

Pattern 2 — adult-services ads. Display names like 「同城上门❤️线下选妃」 ("in-city home visits / offline concubine selection") posting near-identical solicitation text across threads.

Evasion tricks observed (all cheap, all currently working):

• Appending random emoji + a random digit + a random @mention to the fixed spam string
• Occasionally padding with a random Latin word (e.g. Linumusitatissimum)
• Rotating display names and recycled avatars across accounts

Ask

1. Is near-duplicate reply detection applied to CJK content? A sentence repeated verbatim by hundreds of accounts within hours doesn't appear to trip anything.
2. What's the appropriate channel for reporting language-specific spam campaigns like this — issues here, or somewhere else?

Happy to provide more account handles and screenshots.

[Holipori]

Below are screenshots of some examples. These boys are basically everywhere. Here is one post link: https://x.com/dearemon/status/2088488817422631150?s=46

[ByteJoseph]

many accounts posting the exact same sentence verbatim within hours

Not only in Chinese, but in many other languages too. I’ve seen the same thing flooding across many Indian languages. The spammers are exploiting engagement with zero effort. All they want is for people to comment, increase engagement, and make some money or gain followers. X is already shit. They don’t give a shit about how to fix it

it is a problem in X and many other social media platforms, there is no exception. Reply farm accounts are not correctly being detected and many times they lure people to what they want to obtain by spam.
In my view the biggest challenge is to distinguish human behaviour from bot coping human behaviour, and it needs to happen in all platforms.

[cmdr-chara]

many accounts posting the exact same sentence verbatim within hours

Not only in Chinese, but in many other languages too. I’ve seen the same thing flooding across many Indian languages. The spammers are exploiting engagement with zero effort. All they want is for people to comment, increase engagement, and make some money or gain followers. X is already shit. They don’t give a shit about how to fix it

H1B has demolished the platform.

[cmdr-chara]

it is a problem in X and many other social media platforms, there is no exception. Reply farm accounts are not correctly being detected and many times they lure people to what they want to obtain by spam. In my view the biggest challenge is to distinguish human behaviour from bot coping human behaviour, and it needs to happen in all platforms.

If they blocked the third world, often known as India, we would not have these problems.