Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

I was banned from Claude for scaffolding a Claude.md file?

hugodaniel.com

Post reply on HN
Print view View original

451–460 of 664 posts

Re: I was banned from Claude for scaffolding a Claude.md file?

#451

Post by 9rx »

urbandw311er wrote: post #431
PunchyHamster wrote: post #394

We really need some law to stop "you have been banned and we won't even tell you actual reason for it", it's become a plague, made worse with automated systems giving out a ban

Do we though? It’s an important question about liberty - at what point does a business become so large that it’s not allowed to decide who gets to use it? There was a famous case here in the UK of a cake shop that banned a customer for wanting a cake made for a gay wedding because it was contra the owners’ religious beliefs. That was taken all the way up to the Supreme Court IIRC.

Yes, it is not too much to require that that if you offer something to someone that the receiving party is able to have a conversation with you. You can still reject them in the end, but being able to ask the people involved questions is a reasonable expectation — but many of these big tech companies have made it effectively impossible.

If you want to live life as a hermit, good on ya, but then maybe accept that life and don't offer other people stuff?

Re: I was banned from Claude for scaffolding a Claude.md file?

#452

Post by mrkeen »

fluoridation wrote: post #225
mattnewton wrote: post #213

Earlier quoted context omitted.

Because they bought a claude subscription on a personal account and the error message said that they belongs to a "disabled organization" (probably leaking some implementation details).

That's the part I understand. It's the other term that I don't understand.

Anthropic provides an LLM service.

Anthropic banned the author for doing nothing wrong, and called him an organisation for some reason.

In this case, all he lost was access to a service which develops a split personality and starts shouting at itself, until it gets banned, rather than completing a task.

Google also provides access to LLMs.

Google could also ban him for doing nothing wrong, and could refer to him as an organisation, in which case he would lose access to services providing him actual value (e-mail, photos, documents, and phone OS.)

Another possibility is there (which was my first reading before I changed my mind and wrote the above):

Google routes through 3rd-party LLMs as part of its service ("link to a google docs form, with a textbox where I tried to convince some Claude C"). The author does nothing wrong, but the Claude C reading his Google Docs form could start shouting at itself until it gets Google banned, at which point Google's services go down, and the author again loses actually valuable services.

Re: I was banned from Claude for scaffolding a Claude.md file?

#453

Post by svrtknst »

9rx wrote: post #428
Aldipower wrote: post #409

I was recently kicked out from ChatGPT because I wrote "a*hole" in a context where ChatGPT constantly kept repeating nonsense! I find the ban by OpenAI to be very intrusive. Remember, ChatGPT is a machine! And I did not hurt any sentient being with my statement, nor was the GPT chat public. As long as I do not hurt any feeling beings with my thoughts, I can do whatever I want, can't I? After all, as the saying goes,…

> Remember, ChatGPT is a machine! Same goes for HN, yet it does not take kindly to certain expressions either. I suppose the trouble is that machines do not operate without human involvement, so for both HN and ChatGPT there are humans in the loop, and some of those humans are not able to separate strings of text from reality. Silly, sure, but humans are often silly. That is just the nature of the beast.

eh, words are reality. insults are just changes in air pressure but they still hurt, and being constantly subjected to negativity and harsh language would be an unpleasant work environment

Re: I was banned from Claude for scaffolding a Claude.md file?

#454

Post by arghwhat »

sammy2255 wrote: post #450
fuxirheu wrote: post #443

Earlier quoted context omitted.

This can't be real. My chatgpt regularly swears at me. (I told it to in the customisation)

I've been banned from ChatGPT in the past, it gives you a reason but doesn't give the specific chat. And once you're banned you cant look at any of your chats or make a data request

> And once you're banned you cant [..] make a data request

glares in GDPR

Re: I was banned from Claude for scaffolding a Claude.md file?

#455

Post by deaux »

EnPissant wrote: post #434
deaux wrote: post #296

Earlier quoted context omitted.

Codex: Three reasons. I've used all extensively, for multiple months. Main one is that it's ~3 times slower. This is the real dealbreaker, not quality. I can guarantee that if tomorrow we woke up and gpt-5.2-codex became the same speed as 4.5-opus without a change in quality, a huge number of people - not HNers but everyone price sensitive - would switch to Codex because it's so much cheaper per usage. The second one…

Have you tried lower reasoning levels?

Yes and this makes it faster, but still quite a bit slower than Claude Code, and the tool use gap remains. Especially since the comparison for e.g. 5.2 Codex-Low is more like Sonnet than Opus, so that's the speed you're competing with.

Re: I was banned from Claude for scaffolding a Claude.md file?

#456

Post by qcnguy »

cactusplant7374 wrote: post #212
throwup238 wrote: post #195

Earlier quoted context omitted.

You can tell they’re all vibe coded. Claude iOS app, Claude on the web (including Claude Code on the web) and Claude Code are some of the buggiest tools I have ever had to use on a daily basis. I’m including monstrosities like Altium and Solidworks and Vivado in the mix - software that actually does real shit constrained by the laws of physics rather than slinging basic JSON and strings around over HTTP. It’s an utte…

You're right. https://github.com/anthropics/claude-code/issues Codex has less but they also had quite a few outages in December. And I don't think Codex is as popular as Claude Code but that could change.

Don't bother filing issues there. Their issue tracker is a galaxy-sized joke. They automatically close issues after 30 days of inactivity even if they weren't fixed, just to keep the issue count low.

The Reasonable Man might think that an AI company OF ALL COMPANIES would be able to use AI to triage bug tickets and reproduce them, but no! They expect humans to keep wasting their own time reproducing, pinging tickets and correcting Claude when it makes mistakes.

Random example: https://github.com/anthropics/claude-code/issues/12358

First reply from Anthropic: "Found 3 possible duplicate issues: This issue will be automatically closed as a duplicate in 3 days."

User replies, two of the tickets are irrelevant, one didn't help.

Second reply: "This issue has been inactive for 30 days. If the issue is still occurring, please comment to let us know. Otherwise, this issue will be automatically closed in 30 days for housekeeping purposes."

Every ticket I ever filed was auto-closed for inactivity. Complete waste of time. I won't bother filing bugs again.

Re: I was banned from Claude for scaffolding a Claude.md file?

#457

Post by 9rx »

svrtknst wrote: post #453
9rx wrote: post #428

Earlier quoted context omitted.

> Remember, ChatGPT is a machine! Same goes for HN, yet it does not take kindly to certain expressions either. I suppose the trouble is that machines do not operate without human involvement, so for both HN and ChatGPT there are humans in the loop, and some of those humans are not able to separate strings of text from reality. Silly, sure, but humans are often silly. That is just the nature of the beast.

eh, words are reality. insults are just changes in air pressure but they still hurt, and being constantly subjected to negativity and harsh language would be an unpleasant work environment

Words don't hurt. The intent behind those words can. But a machine doesn't carry intent. Trouble is that the irrational humans working as implementation details behind ChatGPT and HN are prone to anthropomorphizing the machine to have intent, which is not reality. Hence why such rules are in place despite being nonsensical.

Re: I was banned from Claude for scaffolding a Claude.md file?

#458

Post by dxdm »

fluoridation wrote: post #373
saghm wrote: post #360

Earlier quoted context omitted.

It seems just as likely to me that they're just using their terminology inconsistently as it is that they're using it consistently but with that egregious amount of ambiguity. The only thing that I'm confident about is that they're communicating in a very confusing way, and that doesn't really give me any strong insight into whether they're being consistent but vague or just plain vague.

Again, I don't agree. If you replace every instance of "non-disabled organization" with just "company", the sentences make sense. There's no need to suppose that the term means anything else, when this interpretation resolves all the outstanding questions satisfactorily and simply.

Just want to say thank you for being patient and rational. Reading your comments in this thread, they're like a soothing bandaid over all this flustered upset.

I wish there were more comments like yours, and fewer people getting upset over words and carrying what feels like resentment into public comments.

Apologies to all for this meta comment, but I'd like to send some public appreciation for this effort.

Re: I was banned from Claude for scaffolding a Claude.md file?

#459

Post by scbrg »

fuxirheu wrote: post #443
Aldipower wrote: post #409

I was recently kicked out from ChatGPT because I wrote "a*hole" in a context where ChatGPT constantly kept repeating nonsense! I find the ban by OpenAI to be very intrusive. Remember, ChatGPT is a machine! And I did not hurt any sentient being with my statement, nor was the GPT chat public. As long as I do not hurt any feeling beings with my thoughts, I can do whatever I want, can't I? After all, as the saying goes,…

This can't be real. My chatgpt regularly swears at me. (I told it to in the customisation)

ChatGPT has too many users for it to be possible to enforce any kind of rules consistently. I have no opinion on whether OP's story is true or not, but the fact that two ChatGPT users claim to have observed conflicting moderation decisions on OpenAI's part really doesn't invalidate either user's claim.

Re: I was banned from Claude for scaffolding a Claude.md file?

#460

Post by Aldipower »

qcnguy wrote: post #449
Aldipower wrote: post #409

I was recently kicked out from ChatGPT because I wrote "a*hole" in a context where ChatGPT constantly kept repeating nonsense! I find the ban by OpenAI to be very intrusive. Remember, ChatGPT is a machine! And I did not hurt any sentient being with my statement, nor was the GPT chat public. As long as I do not hurt any feeling beings with my thoughts, I can do whatever I want, can't I? After all, as the saying goes,…

Can you share that chat?

No, it contains swearwords and sensitive information.
Post reply on HN