Research and data analysis.
📍 Mars
- Likes
- 213K
- Reach
- 13.0M
- Replies
- 6,958
- Reposts
- 39K
Why it's watchedexposing the hidden truth on social networks
402 posts on record from koshercockney. Every one of these five breaks X's TOS:
1. Inauthentic behavior
2. Content spam
3. Bulk and duplicative posting
4. High-volume duplicated replies across many accounts
5. Copypasta
Does X consider this enough to act? t.co/V75srnQLYm
💬 10🔁 45
♥ 209👁 10K
402 replies from @koshercockney.
Only 11 unique replies.
The same reply was posted 102 times in a single day.
Every other repeated reply was posted in under 55 minutes. t.co/r4oreSFAqy
💬 19🔁 89
♥ 414👁 31K
Wait, so let me get this straight:
• March: Loomer says an unnamed CIA source claims Turkey is paying unnamed podcasters. No names, no documents.
• July: A FARA filing comes out listing Kiriakou and “Neil Patel.” It does not mention Tucker, TCN or a payment to either man.
• August: Fischberger says Patel is probably TCN’s co-founder, puts those contacts next to the April Tucker and Kiriakou episode, and Loomer then brings Tucker and Patel into the allegation.
So Loomer gives Fischberger an unnamed claim, Fischberger gives Loomer a Tucker angle, and then the Tucker angle gets treated like support for the unnamed claim?
Am I missing the receipt, or is the receipt just the two posts pointing at each other?
💬 18🔁 34
♥ 277👁 12K
That was the last time he replied to anyone.
This account has been on autopilot for a whole month. 😂 t.co/58wvM5G5UF
💬 9🔁 20
♥ 276👁 6,921
A @StopAntisemites complaint to your boss might look genuine, but the sender never wrote a word.
Some of its posts hand out prewritten complaints as one tap links. Recipient, subject, body already filled in. Sent from a real email. The words aren't theirs.
Here's one example t.co/KKvuFqaKFF
💬 8🔁 78
♥ 342👁 9,317
After depriving 400 people of their livelihood she said
“We’re not in the business of ruining people’s lives… Our goal is to remedy a situation first.” t.co/GvHpeI0yPQ
💬 45🔁 370
♥ 1,722👁 46K
Here are 400 posts from StopAntisemitism.
Red highlights personal labels, accusations, and calls for employers to act against the people named.
Those named include nurses, teachers, students, engineers, and other workers. t.co/tHs6z4x27d
💬 65🔁 337
♥ 1,126👁 114K
Trapped in a car with her dead family, 5 year old Hind Rajab begged a rescue line while bullets rained on her.
His posts, for over a year:
Hamas did it, PIJ did it, crossfire did it, her own foundation faked it. Anyone but the IDF.
No, Max. It was the IDF that killed her. t.co/Z94tIqzfnk
💬 34🔁 343
♥ 1,441👁 74K
Your own footnote is the Community Note.
54% = 29% confirmed foreign + 24% "unknown," mostly Americans on laptops per your fine print.
Your rule confirms a web-only foreigner but calls a web-only American "unknown."
Your largest single foreign origin: Israel, 27 accounts. t.co/hiUsHvCaa1
💬 2🔁 12
♥ 67👁 5,170
Someone with this profile picture liked my post. t.co/kUERqtRtVn
💬 15🔁 22
♥ 413👁 4,712
Still waiting for science to explain this block 😭 t.co/sehQ33bKiZ
💬 9🔁 3
♥ 117👁 2,845
Call out the problems on this platform long enough, and eventually they start treating you like the problem t.co/hu11jomjuy
💬 8🔁 27
♥ 169👁 6,785
Here are some of this week's posts against James Fishback and Dan Bilzerian.
Florida primary week. 20 accounts, 151 posts, a lot of the same attacks. t.co/xe6xrMIYLh
💬 12🔁 35
♥ 187👁 4,571
Here are 288 IDF reprints from @EliAfriatISR.
He sources Hamas 32 times:
hostages, deal terms, Hamas losing.
0 mentions of Gazan casualties.
He is the IDF copy machine. t.co/79YjXPAK0x
💬 7🔁 47
♥ 264👁 3,798
Here are 99 of the 379 hate posts recycled by @ImtiazMadmood.
Same lines, up to 7 times each, most on the exact :00 second.
He writes his pro-Israel posts himself. The hate is a scheduler running copypasta.
When will X fix this? t.co/lpc2nc47Qz
💬 20🔁 88
♥ 373👁 9,722
Here is a clear ToS violation. @diana_bloom_ posted the exact same reply to Reuters 10 times in 143 seconds.
Then she admits the tactic herself: "we flood their comment sections every time." That is platform manipulation. t.co/UBagOl4Szx
💬 31🔁 60
♥ 308👁 8,577
Here are 576 of @EliAfriatISR repetitive posts.
First pic is the same post 144 times.
Each recycled dozens of times, almost all at exactly :00 seconds.
Thats not a person, thats a scheduler running copypasta. Combined, these racked up 149.9 million views.
When will X fix this? t.co/q8Fzoxxix2
💬 93🔁 387
♥ 2,043👁 94K
Next @VividProwess post, dropping in 1 minutes: pro-Israel content, video attached. Screenshot this.
💬 19🔁 25
♥ 268👁 24K
When Elon open sourced X's algorithm, he said X was "actively seeking criticism to improve fairness and quality of algorithmic recommendations."
Here is some.
Every reaction has a price:
Like +0.5
Reply +5
Block -31.2
Mute -58.8
Report -234
But the code also shows the abuse system is not applied equally. X's AI safety pipeline tags targeted insults with FOSNR_ABUSE_INSULTS, which hides the post from recommendations. Accounts labeled High PageRank are skipped. Their posts are not even scanned. They are automatically immune to X's own detection model.
I checked 7,992 of Eitan Fischberger's archived posts. 406 attack named people, "sick freak", "evil witch", "piece of human garbage". No sign any were hidden. They do BETTER than his normal posts.
Are accounts like Fischberger classified as High PageRank? Because the code is clear about everyone else: post these exact insults as a regular account, or reply to him in kind, and the model can scan you, label you, and limit your reach. He may never be scanned at all. One rule for some, another for everyone else. Why?
💬 34🔁 71
♥ 260👁 21K
According to X’s publicly released open source code, its copypasta rule is not applied equally.
Its duplicate text system can give posts a COPYPASTA_SPAM label. But accounts classified as “High PageRank” are skipped by this rule.
Archived posts show:
Dr. Maalouf
902 different content sets reused in 1,922 later posts
Eyal Yakoby
1,365 different content sets reused in 3,412 later posts
VividProwess
294 different content sets reused in 431 later posts within the sample
X does not disclose whether these three are classified as High PageRank. What the archives do show is thousands of reruns across hundreds of recurring sets.
Are these reruns going undetected, or are they being detected and then skipped because the accounts fall under the High PageRank exception?
If so, why?
💬 31🔁 84
♥ 308👁 12K
It was number 6 lol:
“Islam taking over the West.” He recycled the video and still got the country wrong. Copypasta with bad info. t.co/whk2zKb61O
💬 11🔁 44
♥ 260👁 10K
Called when his next post would go up. Nailed it. Called the one after that too.
Then he blocked me. Hopefully the script's off for now. t.co/o1FmBd8XZ2
💬 11🔁 55
♥ 300👁 4,350










































