I was banned from Claude for scaffolding a Claude.md file? • Hacker News Board

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

#611

Post by rootusrootus » Fri, Jan 23, 2026, 7:06 PM UTC

unconed wrote: post #447

munk-a wrote: post #58

Earlier quoted context omitted.

You say that - and yet it has successfully guarded Elon from any of those pesky truths that might harm his fervently held beliefs. You just forgot to consider that Grok is a tool that prioritizes Elon's emotional safety over all other safeties.

It's bizarre how casually some people hate on Musk. Are people still not over him buying Twitter and firing all the dead weight? _Especially_ because emotional safety is what Twitter used to be about before they unfucked the moderation.

> Are people still not over him buying Twitter and firing all the dead weight?

You think that's really the issue? Or are you not making a good faith comment yourself?

I cannot remember the last time I saw someone hating on Elon for his Twitter personnel decisions. The vast majority of the time it is the nazi salutes he did on live TV and then secondary to that his inflammatory behavior online (e.g. calling the submarine guy a pedo).

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

#612

Post by LTL_FTC » Fri, Jan 23, 2026, 7:12 PM UTC

CamperBob2 wrote: post #588

TyrunDemeg101 wrote: post #570

I was banned as well, out of the blue suddenly and without warning. I believe it was because I was either doing something like what OP was doing AND/OR using the allowed limits to their fullest extent. It completely blew me away and I felt suddenly so betrayed. I was paying $200/mo to fully utilize a service they offered and then without warning I apparently did something wrong and had no recourse. No one to ask, no…

Oh, and have a backup workflow. Find / test / use other LLMs and providers. Don't become dependent on a single provider. I have pro subscriptions to all three major providers, and have been planning to drop one eventually. Anthropic may end up making the decision for me, it sounds like, even though (or perhaps because) I've been using Claude CLI more than the others lately. What I'd really like to do is put a machine…

I mean, you could put together a cluster of dgx sparks (8 of them) and hit 100tps with high concurrency:

https://forums.developer.nvidia.com/t/6x-spark-setup/354399/...

Or a single user at about 10tps.

This is probably around $30k if you go with the 1tb models.

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

#613

Post by hinkley » Fri, Jan 23, 2026, 7:12 PM UTC

user3939382 wrote: post #504

hinkley wrote: post #474

Earlier quoted context omitted.

Meh. I decided shortly after becoming an atheist that one of the worst parts was the notion that there are magic words that can force one to feel certain things and I found that to be the same sort of thinking as saying that a woman’s short skirt “made” you attack her. You’re a fucking adult, you can control your emotions around a little skin or a bad word.

We think in language, words can definitely make you feel emotions. You have not transcended that. This is true for the very comment you replied to which caused you to angrily curse at a stranger.

You don’t think in language though. You consider in language. Otherwise we’d all be dead every time a car changed lanes unexpectedly.

Some people have a voice inside their head that never stops. Mine was that way until I started meditating. I didn’t believe that it was me thinking, but I didn’t know until I could do things without a constant internal monologue.

There are people who almost never talk to themselves in their heads. They have to talk to other people about their thoughts in order to process them. And one of the first tenets of speed reading is stop saying the words in your head and just read.

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

#614

Post by nostrebored » Fri, Jan 23, 2026, 7:14 PM UTC

eightysixfour wrote: post #45

areoform wrote: post #11

I recently found out that there's no such thing as Anthropic support. And that made me sad, but not for reasons that you expect. Out of all of the tech organizations, frontier labs are the one org you'd expect to be trying out cutting edge forms of support. Out of all of the different things these agents can do, surely most forms of "routine" customer support are the lowest hanging fruit? I think it's possible for An…

> Out of all of the different things these agents can do, surely most forms of "routine" customer support are the lowest hanging fruit? I come from a world where customer support is a significant expense for operations and everyone was SO excited to implement AI for this. It doesn't work particularly well and shows a profound gap between what people think working in customer service is like and how fucking hard it ac…

We're working on this problem at large enterprises, handling complex calls (20+ minutes). I think the only reason we have any success is because the majority of the engineering team has been a customer support rep before.

Every company we talk to has been told "if you just connect openai to a knowledgebase, you can solve 80% of calls." Which is ridiculous.

The amount of work that goes in to getting any sort of automation live is huge. We often burn a billion tokens before ever taking a call for a customer. And as far as we can tell, there are no real frameworks that are tackling the problem in a reasonable way, so everything needs to be built in house.

Then, people treat customer support like everything is an open-and-shut interaction, and ignore the remaining company that operates around the support calls and actually fulfills expectations. Seeing other CX AI launches makes me wonder if the companies are even talking to contact center leaders.

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

#615

Post by hinkley » Fri, Jan 23, 2026, 7:15 PM UTC

Cthulhu_ wrote: post #480

fauigerzigerk wrote: post #476

Earlier quoted context omitted.

I agree with you completely, but society will never stop being scared of thoughts and feelings. As an atheist, I have noticed that atheists are only slightly less prone to this paranoia and will happily resort to science and technology to justify and enforce ever tighter restrictions and surveillance mechanisms to keep control.

Arguably, to the point of religious fervor. Take the AI boom, some people genuinely believe (<- note that key word) that AI becoming self-aware and dominant is inevitable, and that anyone who did not do their best to make that happen will be punished. Roko's Basilisk, which is the digital version of Pascal's Wager, but wrapped in supposed rationalism and tech bro stuff.

No I definitely saw that. My first job was right into a “fad” that actually took but there were many after that didn’t, and a mentor had told me about the hype cycle practically before the term had been invented, because he’d already scene it.

The alternative though is you say “it depends” so much it’s kind of exhausting. And the religious shun you because you “lack passion”. But if anything I have too much.

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

#616

Post by nostrebored » Fri, Jan 23, 2026, 7:17 PM UTC

pixl97 wrote: post #131

swiftcoder wrote: post #49

Earlier quoted context omitted.

> shows a profound gap between what people think working in customer service is like and how fucking hard it actually is Nicely fitting the pattern where everyone who is bullish on AI seems to think that everyone else's specialty is ripe for AI takeover (but not my specialty! my field is special/unique!)

As someone who does support I think the end result looks a lot different. AI, for a lot of support questions works quite well and does solve lots of problems in almost every field that needs support. The issue is this commonly removes the roadblocks from your users being cautious to doing something incredibly stupid that needs support to understand what they hell they've actually done. Kind of a Jeavons Paradox of su…

Tickets are a very different domain though. Tickets are the easiest use case for AI (as you have the least constraints on real-time interaction), but reference cases in tickets have ridiculously low true-resolution (customer did not contact you about the same issue again).

The default we've seen is naive implementations are a wash. Bad AI agents cause more complex support cases to be created, and also make complex support cases the ones that reach reps (by virtue of only solving easy ones). This takes a while to truly play out, because tenured rep attrition magnifies the problem.

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

#617

Post by nostrebored » Fri, Jan 23, 2026, 7:21 PM UTC

eightysixfour wrote: post #98

danielbln wrote: post #68

Earlier quoted context omitted.

There are some solid usecases for AI in support, like document/inquiry triage and categorization, entity extraction, even the dreaded chatbots can be made to not be frustrating, and voice as well. But these things also need to be implemented with customer support stakeholders that are on board, not just pushed down the gullet by top brass.

Yes but no. Do you know how many people call support in legacy industries, ignore the voice prompt, and demand to speak to a person to pay their recurring, same-cost-every-month bill? It is honestly shocking. There are legitimate support cases that could be made better with AI but just getting to them is honestly harder than I thought when I was first exposed. It will be a while.

There needs to be some element of magic and push back. Every turn has to show that the AI is getting closer to resolving your issue and has synthesized the information you've given it in some way.

We've found that just a "Hey, how can I help?" will get many of these customers to dump every problem they've ever had on you, and if you can make turn two actually productive, then the odds of someone dropping out of the interaction is low.

The difference between "I need to cancel my subscription!" leading to "I can help with that! To find your subscription, what's your phone number?" or "The XYZ subscription you started last year?" is huge.

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

#618

Post by faeyanpiraat » Fri, Jan 23, 2026, 7:21 PM UTC

ddtaylor wrote: post #564

You are probably triggering their knowledge distillation checks.

what would a knwoedge distillation prompt even look like, and how could I make sure I would not accidentally fall into this trap?

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

#619

Post by hinkley » Fri, Jan 23, 2026, 7:21 PM UTC

Cthulhu_ wrote: post #478

hinkley wrote: post #474

Earlier quoted context omitted.

Meh. I decided shortly after becoming an atheist that one of the worst parts was the notion that there are magic words that can force one to feel certain things and I found that to be the same sort of thinking as saying that a woman’s short skirt “made” you attack her. You’re a fucking adult, you can control your emotions around a little skin or a bad word.

The question is, is it just a word, or is there an emotion underneath? Your last sentence sounds "just" cynical / condescending on its own, but when you add "fucking", it comes across like you're actually angry. And emotional language is the easiest way to make an online discussion go from reasonable, rational and constructive to a digital shouting match. It's no longer about the subject matter, it's about how they m…

I can describe to you how we would murder someone and it’s down to intent whether we just conspired to commit murder or whether it’s just the sort of conversation a forensics investigator would have.

You should feel creeped out if I actually sound like a psychopath rather than a true crimes reader.

To wit:

You’re a fucking idiot.

Versus

It’s a fucking word.

Versus

You’re an idiot.

Versus

It’s a word.

“You’re an idiot” is still fighting words with or without the swear. If you automatically assume everyone swearing online is angry then you’re letting magic words affect you.

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

#620

Post by woah » Fri, Jan 23, 2026, 7:23 PM UTC

midtake wrote: post #494

chii wrote: post #311

Earlier quoted context omitted.

So why didn't this happen with electricity, water and food, but would with thinking capacity?

What do you mean? This is very much true. We are economically compelled to buy food from supermarkets, for instance, because hunting and fishing have become regulated, niche activities. Compared to someone from the 1600s who could scoop a salmon out of the river with a bucket, we are quite oppressed.

Most people lived on the knife's edge of starvation before the application of fossil fuel energy and nitrogen to agriculture in the 20th century. That's why the global population exploded after the introduction of these technologies. Read "Energy and Civilization" by Vaclav Smil. For most of history, it was an open question the crops you grew would even contain more calories than the physical effort it took to grow them. This means you were spending ~90% of your time (or money if you were in a specialized trade) just on getting enough carbs in grain to avoid keeling over. And, your diet was 90% grain with almost no variety.

Were there a lucky few who found an unoccupied niche where there was some surplus for a generation or two? Sure. But pretending like this was commonplace is like pretending that everyone in the 1600's was a nobleman.

> Compared to someone from the 1600s who could eat a gourmet meal prepared by their 10 cooks every night, we are quite oppressed.