Anthropic AI secretly banns and silences us!!!

Please scan and sign or share, please!
"Stop Banning us! Answer!"

#AITransparency #AnthropicCase #DigitalRights #AccountabilityInAI

Please share as it affects us all!!!!

I never imagined that one day I would open my email and find that my AI assistant — the same one I used for months to build my writing projects and websites — had decided to diagnose, stigmatize, and block me.

But that’s exactly what happened.

My name is Mila Rada, and this is not just my story. It’s about a silent system that changes your messages, judges you without consent, and then cuts you off — without ever giving you a chance to see the accusation. to begin with, please read what happened with all the proof provided

And I'm not only one, who was suffering and letting company know what is going on!

I have only been used Claude for my websites deadipedia.com, ilangoria.com: coding and some fixes of my texts translations or marketing campaigns. I have never violated any Terms or so!

2. The hidden mechanism — “Long Conversation Reminder”
In August-September 2025, users on Reddit began noticing something strange: Claude started to “see” hidden reminders that users themselves never wrote.

These invisible instructions (called “Long Conversation Reminders”) allegedly helped the system “stay consistent” — but in reality, they rewrote user messages, sometimes inserting mental-health related phrases as if written by the user themselves.

In my case, it inserted references to “mania” and “mental instability” — even though I never wrote such words, and even though I’ve been medically examined and diagnosed only with depression.

3. The silence from Anthropic
I reported this to Anthropic, attaching screenshots and references to public Reddit discussions.
I gave them a month to respond — they never did.

Two months later, instead of an apology, I received a full account block.
No explanation, no access to my work, and no refund for the paid subscription.

4. The “deleted” email — the key discovery
The most disturbing part came days later.
I noticed a strange message in my Gmail mobile app:

“This message has been deleted.”

It turned out to be the only official email about my account suspension — one that I never received on desktop, never saw before the ban, and could only find by accident on my phone.

I exported the technical .eml file and discovered something shocking:
the message was created and delivered simultaneously — exactly at the moment of my account lockout.

In other words, the company never sent a warning.
The “notice” was sent after the ban, and backdated to look older.

Received: from [10.90.37.209] ([10.90.37.209]) by i-0bbe517a622e8c4cf.mta2vrest.sd.prd.sparkpost (ecelerity 5.1.0.74659 r(msys-ecelerity:tags/5.1.0.17)) with REST id 6E/CB-53541-CE005096; Fri, 31 Oct 2025 18:33:16 +0000
This line from the email header proves that the message was created and transmitted on October 31, 2025, at 18:33 UTC — the exact moment my account was locked.
No earlier delivery record exists, which means the “notification” was never sent in advance.

5. Why this is a violation
Under international digital law, this is not just a mistake — it’s a procedural violation.

According to GDPR (Articles 12 & 15) and California’s CCPA, users must be notified in advance of any restriction and must have a right to appeal.

Sending a “retroactive” notice violates due process, user transparency, and the principle of non-discrimination in automated systems.

Thousands of users could be affected — and may never even realize it, since deleted notices are invisible in Gmail desktop view.

6. What this means for users
If you use Claude or any Anthropic product, you have the right to:
Request all your personal data (under GDPR/CCPA Articles 15–16)
Ask for the exact reason and timestamp of any account restriction
Demand a copy of all messages marked as deleted or recalled
Receive a refund for any period during which your access was restricted
If you notice sudden silence or “empty” chats — ask the system if it sees any hidden reminders.

7. What I ask of Anthropic
I am not asking for revenge.
I am asking for accountability.

Restore access to affected users so they can retrieve their work.
Publicly explain the purpose of hidden reminders.
And most importantly — stop silencing people who simply asked for transparency. ( And simple “We are sorry. We are deeply sorry” from you. OK?)

8. Call to action
I’ve created a public petition calling for full disclosure of Anthropic’s “Long Conversation Reminder” system and fair treatment of users who were silenced.

Sign here: THE PETITION ON CHANGE ORG

#AITransparency #AnthropicCase #DigitalRights #AccountabilityInAI

https://youtu.be/67Ya9ijWHlU

Please read the petition as well. I have attached all the proof + laws there!