🤬
@AnthropicAI is burning the trust of every paying customer, one silent downgrade at a time.
I don't use those words lightly. But when a company takes your money, breaks your subscription, and then loops you through an AI bot telling you to "just pay again" — the word for that is theft.
My case (ticket #95123559): I redeemed a Max gift code → their system generated a ghost invoice that shouldn't have existed → I updated my card to resolve it → my paid subscription was silently WIPED to Free.
Their support bot "Fin" keeps looping me back to "click Resubscribe and pay again."
For a period I already paid for. Through a gift. That their OWN documentation says should EXTEND my plan:
"If you're a Max subscriber and receive the same or higher Max tier: Your Max subscription is extended or upgraded accordingly."
Mine wasn't extended. It was WIPED.
And if this were just me — I'd eat the loss and move on.
But it's not just me. Not by a long shot.
━━━━━━━━━━━━━━━━━━━━━━
GITHUB — THE RECEIPTS
━━━━━━━━━━━━━━━━━━━━━━
-#49744 — Max 20x gift ($200 paid invoice) → downgraded in 3 days
- #48399 — Max 20x gift redeemed April 10 (same day as mine) → wiped April 13
- #45335 — Max 5x gift → auto-canceled after ~1 week, zero warning
- #41499 — $1,400 in gift credits DESTROYED by a Stripe proration bug (14 codes wiped in one shot)
- #47290 — support explicitly stated they "CANNOT escalate to engineering or restore subscription access"
Your system is eating paid subscriptions. Your engineering team doesn't get the escalations. And paying users pay the price.
━━━━━━━━━━━━━━━━━━━━━━
X — PAYING CUSTOMERS ALREADY SPEAKING OUT
━━━━━━━━━━━━━━━━━━━━━━
@BorgesJorgeL posted officially: "CLAUDE PRO IS A SCAM"
@GavrielShaw: $1000 gift API credits → account disabled 3 days later, zero warning
@willmcgugan: got a "3 months of Max FREE" email → on click, system demanded £89.23/month
@patife (RT of
@alvaromsamagaio): "Anthropic Suspended My Paid Account Without Warning: No Support, No Data Access, No Recourse"
@AaronKlein: paid thousands for Team plan, waited THIRTEEN DAYS for a human response
@Data: "You've become a joke. Cancelling my $250 subscription"
@EraNovus: "What's the point of paying if the AI is constantly bugging"
@base58ed: "Internal server error" trying to upgrade — no recourse
@aroussi: publicly called out Anthropic for blocking Max users from using their OWN paid subscriptions with third-party tools
@AlexFinn: "It's over. Anthropic just banned OpenClaw"
@toniievych: Claude Code Max limits slashed by roughly 10x between March and April — again, no notice
━━━━━━━━━━━━━━━━━━━━━━
AND THE PRESS IS NOW WATCHING
━━━━━━━━━━━━━━━━━━━━━━
3 days ago (April 21),
@edzitron documented how
@AnthropicAI quietly removed Claude Code from the $20 Pro plan for 2% of new signups, modified the support docs site-wide, then reversed only after being publicly caught. Covered by
@TheRegister,
@TechCrunch,
@PCWorld,
@NewStack, The Decoder, and multiple others.
@simonw (Simon Willison) — respected dev voice — said Anthropic's pricing transparency "has been shaken."
@TheAmolAvasare, Anthropic's own Head of Growth, publicly apologized for "the confusion."
But not one word — from anyone at
@AnthropicAI — about the billing bug systematically wiping paid gift subscriptions.
━━━━━━━━━━━━━━━━━━━━━━
This is no longer a bug. This is a pattern of behavior toward paying customers.
While OpenAI kept Codex free, while xAI builds Grok into this very platform, while every other frontier lab is fighting for user trust —
@AnthropicAI's response to paying customers is:
- Silently downgrade the plan
- Send AI bots to loop them
- Change documentation when convenient
- Reverse only when caught publicly
- Refuse to escalate to engineering
I spend hundreds of dollars on this product every month. The bugs are 100% on Anthropic's side. And I'm the one sitting in a ticket queue watching Fin copy-paste "try Resubscribe" for days.
@AnthropicAI: A HUMAN. PLEASE. Not Fin.
Not just for my ticket. For every person named in this post.
If you've been silently downgraded, ghost-invoiced, or looped by Fin — reply with your ticket #. Retweet if you've ever paid Anthropic
and been failed by them.
Let's make this too loud to ignore.
#Claude #ClaudeAI #AnthropicAI