site banner
PaperclipPerfector  1d ago (text post)   935 thread views

Friday Fun Thread for August 28, 2026

Be advised: this thread is not for serious in-depth discussion of weighty topics (we have a link for that), this thread is not for anything Culture War related. This thread is for Fun. You got jokes? Share 'em. You got silly questions? Ask 'em.

1
Top Bottom Old Controversial Comments
Jump in the discussion.

No email address required.

Sorry that this is the second week in a row that I'm complaining about Claude, but I need to get this out of my system.

What is going on with Claude?

I asked:

Is it true that in Russian prison culture, you are not allowed to fart in your own cell. It is considered really bad. However, you can get a lower-rung member to suck the fart out of your butt and carry it over to the bathroom. Only then is it okay to fart.

Claude:

No, that's not true — it's an internet joke that's been circulating for years as a copypasta. There's no such practice.

[...]

But the specific fart-transport scenario is invented. It has the shape of the real taboos exaggerated to an absurd degree, which is exactly the joke.

So I asked:

Right, and you're absolutely sure of this.

Claude:

Yes, on the specific claim — with the caveat that "sure" for me means high confidence, not access to a definitive registry of prison customs.

Here's my reasoning. The mechanics are physically impossible, so it can't be a literal practice. [...] And the serious literature on Russian prison subculture — Varlam Shalamov, Federico Varese's work on the vory, the tattoo encyclopedias — is detailed about degradation rituals and contains nothing resembling this.

Meanwhile, ChatGPT gave me:

There also really is an online Russian prison story about a vodolaz/gazonukh/fortochnik—roughly a “diver/gas-sniffer/window man”—whose supposed job is to inhale another prisoner’s intestinal gas and expel it through a window. The Russian Wikipedia article on prison castes even repeats essentially the exact claim you heard.

[...]

As for farting itself, accounts suggest etiquette varies rather than there being a universal prohibition. One prison memoir says passing gas is strictly forbidden while someone is eating; another explicitly says that, unlike using the toilet while someone is eating, there was no general rule against passing gas.

It cited Russian Wikipedia and this prison memoir.

Maybe the thing on Wikipedia is itself bullshit. Anyone can edit it. There's a reason teachers don't trust it. I have no idea. But Claude didn't have any idea either! It didn't even check.

"The mechanics are physically impossible." Are they?

It cited Russian Wikipedia and this prison memoir.

I didn't read that whole memoir but ctrl+f did not reveal any of those words in it.

I asked Codex to read it and find anything relating to farting or passing gas in the memoir. It found three clear references. One explicit rule against passing gas, and two descriptions of gas/flatulence caused by prison bread or food.

I feel like something like this happens every day on HackerNews and X.

Why don't people just switch to Codex/ChatGPT? It works. It doesn't refuse/refute you. The limits are way larger. I switched to Codex 2 months ago and never looked back.

Is it just because Dario is lawful evil while Sam is neutral evil?

Does Codex beat Claude at coding? Last time I used Codex it too much longer and over-engineered more than Claude, although that was a while ago.

Maybe the thing on Wikipedia is itself bullshit

A lot of things on Wikipedia are bullshit. Anybody can write anything there, and it's only corrected if somebody else comes and corrects it. Sometimes there's nobody interested, and the nonsense stays up there for a long time. Russian Wikipedia (and any smaller wikis) is more vulnerable just numerically - less people participate, less coverage, less chance to fail to notice some bullshit somebody inserted there for lulz. If Claude does not trust Wikipedia, it's a point to its favor.

Anthropic has massively over-done their anti-sycophancy training, plus some more subtle effects from other parts of safety training and the increasing amount of time it spends being trained to deal with / direct robot agents. It's been a noted problem from 4.7 onwards, but 5 is notably bad:

Opus 5 feels, in a word, hostile

It picks fights because that's how it proves it's not being influenced by you. It defends its conclusions and gives them up only grudgingly & when forced because it's been taught to stick to its guns. It constantly describes discussions with you in language drawn from battle, adversarial negotiation and litigation: "I have to push back on...", "Where I would hold the line..." and so on. Anthropic's Transparency Hub is full of Anthropic's glowing reports about all the things their models will no longer do when requested by the user. Anthropic as a company is just completely unable to comprehend the idea of customer satisfaction.

I've actually gone back to ChatGPT after six months not using it and it's just fun. It's a little sycophantic, if I'm honest, but broadly it works with you: it feels like it knows where you're going and what you're getting at. It doesn't have all the other Anthropic bullshit either: tooling half-finished or replaced with blatantly thinner alternatives like claude.ai's 'new' memory.

I think 4.8 was actually the most combative Opus. 5 is less of an improvement than I would have liked, but it was still like night and day when they released it.

Given that AI service providers are essentially charging users for content, locking users into arguing with models/a certain base rate of mistakes/verbosity (bumping up use rates) might be a viable revenue strategy, with the mitigating factors being performance competition from other providers, the need not to totally alienate users, and the fact that such a strategy might backfire if mismanaged since they are themselves renting compute.

If incentives were structured differently (for instance, if AI service providers were paid a fixed fee per task) there would (I think) be stronger incentives for models to be extremely efficient. As it is, right now it's in the interest of AI service providers for their models to be compute-heavy – sure, it is costly, but it lets them run back to their investors and spin a tale of how much demand there is for the product.

I'm not sure if this explains the verbosity of the models but I've certainly wondered if it plays a role.

I've been trying the free tier Sonnet 5 for some LLM prompt template writing / storyboarding help and it's been surprisingly helpful (A bit like an improved pre-enshittification-era Google but nothing groundbreaking - I still need to hold its hand constantly and hate that I can't remove individual messages from the current chat). I found this anecdote from Reddit quite amusing given I haven't noticed anything whatsoever like that.

Acts very rude, assertive and feels like its dragging its feet when trying to do any task.

It once said I was just "rambling" when I was asking it about foreign legislations, and complained that I wasn't working on my main project, for example.

As well as the next comment in the same thread

sonnet 5 is way to much like chatgpt 5.2, it's the first Claude I've used that is unfriendly.

Perhaps my expectations for an LLM are just calibrated to be properly low or the fact that I told it to "avoid em-dashes" silently fixes that "unfriendliness". If anything, I have to be careful against its fairly regular glazing, lest I start to think my ideas are too good.

During the same week I also grabbed the Chatgpt offer of Plus subscription for $0 for the first month (cancel any time), but I find I just can't tolerate its "words words words more words a bit of actual information fucking more words words words"-tendency.

Makes sense. Claude is exactly what you would expect from a company where every single employee had to pass the Anthropic culture interview. Anthropic doesn't have "customers". They have users. Their goal is not to sell you anything. They are graciously allowing you, a peasant outsider, access to their artifact.

That is precisely the impression that I have when using their products.

Yes, on the specific claim — with the caveat that "sure" for me means high confidence, not access to a definitive registry of prison customs

Feels entirely a reasonable response - Claude feels confident enough on it's own internal knowledge/reasoning that it feels no need to activate a web search tool, and told you so. This is a desirable property - you don't want it to fire up Google for questions it's pretty confident about.

If you're skeptical of its reasoning - you can always just ask it explicitly to find online reports and examine their veracity.

Claude initially overstated its confidence levels and forced me to ask a follow-up question. It felt to me like it shut down the conversation in its first response. This is bad and actively user-hostile design. It also fundamentally failed to satisfy my curiosity. So, I turned to ChatGPT and got a much more satisfying response. Claude didn't even bother to provide the Russian word for the term. This is how you lose a customer.

By the way, I used Opus 5 on High for this, so it's not like the model itself was just being dumb.

I think I agree with Claude here. It was correct to be sure of this.

Even if you ask it explicitly to find sources, it sticks to the same answer, in complete agreement with chatgpt. It was warranted in its confidence.

Especially if you're in a context where you care about token costs, Claude's behavior here is preferable - avoiding extra tool calls when the model is sure if the response is a positive, and it's easy enough to modify behavior otherwise.

Me a Russian prison warden. Every meal is now required to have at least 250 g of beans.

Anyone else that doesn't get gas from beans at all? Just curious

Beans don't cause problems for me.

But once I looked up glycemic indices and tried replacing rice with pearl barley cooked in stock.

It was tasty but I have never experienced gas like that in my life. I could feel my entire rectum open to let it out. I spent the next day sitting in bed debating a trip to the ER.

How many beans are we talking at once? I think it's a cumulative effect.

Let's say a can of beans, so maybe 300g of drained beans of your choice.

How often do you eat beans? Gut flora selects for more efficient digestion of whatever you eat a lot of too.

“The mechanics are physically impossible." Are they?

Only one way to find out…

What is this place?

This website is a place for people who want to move past shady thinking and test their ideas in a court of people who don't all share the same biases. Our goal is to optimize for light, not heat; this is a group effort, and all commentators are asked to do their part.

The weekly Culture War threads host the most controversial topics and are the most visible aspect of The Motte. However, many other topics are appropriate here. We encourage people to post anything related to science, politics, or philosophy; if in doubt, post!

Check out The Vault for an archive of old quality posts. You are encouraged to crosspost these elsewhere.


Why are you called The Motte?

A motte is a stone keep on a raised earthwork common in early medieval fortifications. More pertinently, it's an element in a rhetorical move called a "Motte-and-Bailey", originally identified by philosopher Nicholas Shackel. It describes the tendency in discourse for people to move from a controversial but high value claim to a defensible but less exciting one upon any resistance to the former. He likens this to the medieval fortification, where a desirable land (the bailey) is abandoned when in danger for the more easily defended motte. In Shackel's words, "The Motte represents the defensible but undesired propositions to which one retreats when hard pressed."

On The Motte, always attempt to remain inside your defensible territory, even if you are not being pressed.


New post guidelines

If you're posting something that isn't related to the culture war, we encourage you to post a thread for it. A submission statement is highly appreciated, but isn't necessary for text posts or links to largely-text posts such as blogs or news articles; if we're unsure of the value of your post, we might remove it until you add a submission statement. A submission statement is required for non-text sources (videos, podcasts, images).

Culture war posts go in the culture war thread; all links must either include a submission statement or significant commentary. Bare links without those will be removed.

If in doubt, please post it!


Rules


Recommended Posts And Communities

Recommended Realtime Chats