frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Open in hackernews

Anthropic disabled my account after payment cancer patient/medical data trapped

82•marichala•2h ago
*Title:* Anthropic disabled my Max account after charging me $106.60 — I’m a cancer patient and my medical documentation is trapped inside

*Post:*

On January 16th, 2025, Anthropic charged me $106.60 for my Max subscription and disabled my account in the same moment. I believe I was flagged because I’m a long-term guest at a Marriott hotel using shared WiFi.

I’m not a bot. I’m not running scripts. I’m not violating terms of service. I’m a 41-year-old woman with MDS (myelodysplastic syndrome) that has converted to leukemia. I’m facing a bone marrow transplant I may not survive.

For months, Claude has been the only thing that actually helped me navigate a medical system that failed me for over a decade. It helped me organize 11 years of medical records, track my labs, draft insurance appeals, and write letters to doctors who wouldn’t listen. That work is what finally got someone to take me seriously.

My last prompt before the ban was asking Claude to help me interpret my October bloodwork. That’s it. That’s the “violation.”

That chat history is my medical documentation. I need it to continue advocating for my care when I’m too sick to remember what happened, what was said, what was missed. Without it, I lose years of work at the worst possible moment.

I have done everything I was supposed to do:

- Submitted appeals through the official form - Emailed support@anthropic.com - Emailed usersafety@anthropic.com - Filed a complaint with the California Attorney General - Filed a complaint with the FTC - DM’d and tweeted at Daniela Amodei, Dario Amodei, Amanda Askell, Jan Leike, and Mike Krieger - Posted in the official Claude Discord #community-help channel

Every response has been automated. The support bot offered me a human agent. I accepted. No human ever came.

It can’t be a content violation — my last prompt was about bloodwork. It can’t be a payment issue — the charge went through. The only explanation is their system flagged my IP because I’m on shared hotel WiFi, and no human has reviewed it since.

I am asking for one of two things:

1. Restore my account 1. Export my complete chat history and send it to me

Anthropic talks constantly about building AI that helps people. Claude helped me. It helped me fight for my life. Now I can’t get a single human being at the company to look at my case.

If anyone here works at Anthropic or knows someone who does, I would be grateful for any help.Thank you.

Comments

MattGaiser•1h ago
I wish companies had a pay $50 to speak to a human option if need be.

No spamming abuser will pay that, but it should easily cover the cost of an overseas support agent to handle edge cases.

LtWorf•1h ago
You know they'd put you in touch with someone who has no clue about anything.
p4coder•1h ago
I think what we really need is Internet user's Bill of Rights. The power and information asymmetry is too great to obtain fair treatment for the user.
nottorp•1h ago
So they will hire a "support agent" that costs $30 and leave their service as buggy as bearable so they make those extra $20 more often.
asmor•1h ago
The problem with rate limiting with money is always that it's also going to be too much for someone else to pay (or too little for others).

I'm so glad ICANN banned silent auctions for the next round of gTLDs.

rschiavone•1h ago
That would cause companies to provide the shittiest possible service in order to monetize their support. And they would still outsource the support for $1/hr.
BrenBarn•52m ago
I wish governments had a "pay $100 million if you don't have a human option for all your customers" option.
nobodywillobsrv•1h ago
Anthropic should do a full inquiry into this and fire people for it. No questions asked. Fire them. And then reply with automated bits.
euazOn•1h ago
I don't know. If this is a real thing, then I'm really sorry for what happened to you.

But this whole post seems a bit fishy to me. Brand new account, and it starts with "Title:" and "Post:", the whole thing being obviously entirely AI generated, and a few other signs.

IshKebab•1h ago
This is one really great thing about the GDPR - you could just file a personal information access request and they legally have to give you all your data.

(Sorry that doesn't help you.)

edarchis•1h ago
I got banned for asking about the yfinance python module. They had an "appeal" but it was a Google Form that probably nobody ever looks at.

My recommendation would be to get in touch with their DPO (Data Protection Officer) and invoke the GDPR rule that you have the right to 1. have an explanation as to why an automated decision was made about you, 2. ask for a human review. You are out of the GDPR scope but the legal contact might not bother checking and just restore your account. Getting your data is also a right under GDPR but getting your account back would be a better option. I wouldn't mention this or they'll jump on it.

user_7832•1h ago
Sorry to hear. (Assuming you don't get help from here) The "best" solution very likely involves Anthropic's legal team.

Depending on how much time, energy, and money you have, you can either draft a (simple) email to their customer support CC'ing legal (with ChatGPT/Gemini thinking's help), or ideally get a lawyer to do the same (easier if one of your friends is one).

If you don't get a response to the email, send a certified letter, ideally with a lawyer's help.

I'm not sure on what legal grounds you could sue (I'm sure a good lawyer would find a few - not providing your data for a Cali resident seems an obvious one to me) - but getting legal involved is often enough to "wake up" large cos into getting a human in the loop. A certified letter needs acknowledgement of the recepient, again mandating a human.

Best of luck.

smca•1h ago
(I work at Anthropic) I’ll exhaust every option to have this fixed for you.

If you can pass along at least some way to get in touch (link your Twitter account or similar) that would be helpful.

jwrallie•1h ago
If this is real, best of luck to get your data back.

So many companies are operating with basically no support nowadays, so for the end user you just hope everything works perfectly (until it doesn't).

I don't even blame the bots, I've had human interactions in the past where, while they could perfectly understand my problems, they couldn't do anything because they worked on a script and have limited agency. It is basically the same flow, you just talk until it either escalates to someone that does have the ability to help or it loops back to the beginning and you try again.

A decentralized peer-to-peer messaging application that operates over Bluetooth

https://bitchat.free/
131•no_creativity_•2h ago•58 comments

MTOTP: Wouldn't it be nice if you were the 2FA device?

https://github.com/VBranimir/mTOTP/tree/develop
36•brna-2•1h ago•28 comments

Gaussian Splatting – A$AP Rocky "Helicopter" music video

https://radiancefields.com/a-ap-rocky-releases-helicopter-music-video-featuring-gaussian-splatting
622•ChrisArchitect•16h ago•187 comments

Provide agents with automated feedback

https://banay.me/dont-waste-your-backpressure/
130•ghuntley•1d ago•56 comments

Dead Internet Theory

https://kudmitry.com/articles/dead-internet-theory/
305•skwee357•13h ago•368 comments

Flux 2 Klein pure C inference

https://github.com/antirez/flux2.c
331•antirez•15h ago•119 comments

Show HN: I quit coding years ago. AI brought me back

https://calquio.com/finance/compound-interest
119•ivcatcher•8h ago•105 comments

Self Sanitizing Door Handle

https://www.jamesdysonaward.org/en-US/2019/project/self-sanitizing-door-handle/
9•rendaw•3d ago•8 comments

Fil-Qt: A Qt Base build with Fil-C experience

https://git.qt.io/cradam/fil-qt
93•pjmlp•2d ago•49 comments

A Social Filesystem

https://overreacted.io/a-social-filesystem/
397•icy•1d ago•161 comments

AVX-512: First Impressions on Performance and Programmability

https://shihab-shahriar.github.io//blog/2026/AVX-512-First-Impressions-on-Performance-and-Program...
63•shihab•5d ago•21 comments

The Code-Only Agent

https://rijnard.com/blog/the-code-only-agent
71•emersonmacro•7h ago•31 comments

Show HN: AWS-doctor – A terminal-based AWS health check and cost optimizer in Go

https://github.com/elC0mpa/aws-doctor
26•elC0mpa•5h ago•11 comments

Gas Town Decoded

https://www.alilleybrinker.com/mini/gas-town-decoded/
135•alilleybrinker•4d ago•112 comments

Gladys West's vital contributions to GPS technology

https://en.wikipedia.org/wiki/Gladys_West
17•hackernj•2d ago•2 comments

Simulating the Ladybug Clock Puzzle

https://austinhenley.com/blog/ladybugclock.html
26•azhenley•1d ago•4 comments

Show HN: Pdfwithlove – PDF tools that run 100% locally (no uploads, no back end)

https://pdfwithlove.netlify.app
153•pratik227•4h ago•101 comments

Astrophotography visibility plotting and planning tool

https://airmass.org/
28•NKosmatos•3d ago•5 comments

Command-line Tools can be 235x Faster than your Hadoop Cluster (2014)

https://adamdrake.com/command-line-tools-can-be-235x-faster-than-your-hadoop-cluster.html
373•tosh•1d ago•247 comments

Show HN: Dock – Slack minus the bloat, tax, and 90-day memory loss

https://getdock.io/
134•yadavrh•13h ago•116 comments

Show HN: Beats, a web-based drum machine

https://beats.lasagna.pizza
89•kinduff•12h ago•25 comments

Using proxies to hide secrets from Claude Code

https://www.joinformal.com/blog/using-proxies-to-hide-secrets-from-claude-code/
78•drewgregory•5d ago•24 comments

High-speed train collision in Spain kills at least 39

https://www.bbc.com/news/articles/cedw6ylpynyo
134•akyuu•9h ago•112 comments

Experiments with Kafka's head-of-line blocking (2023)

https://www.artur-rodrigues.com/tech/2023/03/21/kafka-head-of-line-blocking.html
11•teleforce•5d ago•0 comments

How to wrangle non-deterministic AI outputs into conventional software? (2025)

https://www.domainlanguage.com/articles/ai-components-deterministic-system/
34•druther•3d ago•27 comments

Sins of the Children

https://asteriskmag.com/issues/07/sins-of-the-children
148•maxall4•16h ago•71 comments

Police Invested Millions in Shadowy Phone-Tracking Software Won't Say How Used

https://www.texasobserver.org/texas-police-invest-tangles-sheriff-surveillance/
311•nobody9999•12h ago•88 comments

Show HN: Lume 0.2 – Build and Run macOS VMs with unattended setup

https://cua.ai/docs/lume/guide/getting-started/introduction
125•frabonacci•15h ago•35 comments

Poking holes into bytecode with peephole optimisations

https://xnacly.me/posts/2026/purple-garden-first-optimisations/
26•xnacly•4d ago•0 comments

ASCII characters are not pixels: a deep dive into ASCII rendering

https://alexharri.com/blog/ascii-rendering
1250•alexharri•1d ago•134 comments