Claudette: Make Claude stop talking like a BuzzFeed article • Hacker News Board

Re: Claudette: Make Claude stop talking like a BuzzFeed article

#221

Post by unclebucknasty » Sat, Aug 22, 2026, 3:49 PM UTC

bonoboTP wrote: post #216

multjoy wrote: post #186

Earlier quoted context omitted.

>It's like it has no conception This is the point.

It's an empty point and a boring one and leads nowhere. It means nothing, but sounds deep. It's so shallow, any taxi driver and sociologist can come up with it and has, already in December 2022.

>It's an empty point and a boring one and leads nowhere.

But, is it true?

>It means nothing...shallow...

Ironically, the entirety of your comment just repeats that the GP comment means nothing. There is no further explanation or "depth".

Re: Claudette: Make Claude stop talking like a BuzzFeed article

#222

Post by bonoboTP » Sat, Aug 22, 2026, 3:49 PM UTC

bhelx wrote: post #114

graemep wrote: post #76

Earlier quoted context omitted.

Claude not only writes verbose comments, it also writes comments about how things used to work when refactoring. That might have a place in version control comments, but not in the code.

This speaks to the general problem with using LLMs for writing. The audience they are writing for us you, but you're trying to write for a totally different audience. In code, this manifests as comments in the code that are hyperspecific to the conversation you are having, and not the long term benefit of having those comments in the code. I see this in docs a lot. I've been reading a lot of docs these days where it…

Yes. Though adding something like what you wrote here into the instructions does help. Especially combined with a fresh session.

Re: Claudette: Make Claude stop talking like a BuzzFeed article

#223

Post by YuriNiyazov » Sat, Aug 22, 2026, 3:54 PM UTC

sunnybeetroot wrote: post #172

YuriNiyazov wrote: post #68

Earlier quoted context omitted.

It's easiest to explain this while anthropomorphizing the model, I know some folks here hate that, sorry about that. I heard an interesting diagnosis for why Claude does this: the output is a compressed version of its thought traces, very dense because the model is under pressure to use as few tokens as it can and to pack as much (for accuracy) of its concepts into the output. One of the reasons that "don't do X" typ…

Hey Yuri, slight tangent but how did you get to be an investor on those private companies as per your LinkedIn?

Wefunder

Re: Claudette: Make Claude stop talking like a BuzzFeed article

#224

Post by ksajadi » Sat, Aug 22, 2026, 4:16 PM UTC

But it will remove the load bearing seams for the three things it might want to highlight before the gate is open.

Re: Claudette: Make Claude stop talking like a BuzzFeed article

#225

Post by lucisferre » Sat, Aug 22, 2026, 4:21 PM UTC

nrmitchi wrote: post #84

datakan wrote: post #9

Has Anthropic said anything about how or why Claude writes the way it does? So many people hate it, seems like they need to do some damage control there. I haven't had the same problems others have but I'm also not a heavy user of it.

I do not have evidence or data that supports this. It is only my thought. Claude, since Opus 5, speaks more and more like a wannabe-thought-leader pontificating on social media for engagement. Everything is a bait-then-switch, or a multi-post story format. The "engagement" that works well for social media makes actual work extremely frustrating. My unsupported belief is that this is caused by an obnoxious number of p…

Evidence or not, this is probably the most satisfying explanation I've heard for this behaviour.

My own suspicion is that LLMs are not becoming more general-purpose over time, as these companies had hoped, and ongoing development in that direction is stalling. They will likely need to move in the direction of more specialization and train LLMs for specific use-cases. However, this would also be admitting that they are not on a direct path to AGI.

Re: Claudette: Make Claude stop talking like a BuzzFeed article

#226

Post by irpap » Sat, Aug 22, 2026, 4:31 PM UTC

PeterStuer wrote: post #73

Isn't it absurd that everyone is installing these bandaids or rolling their own? Wtf is Anthropic smoking?

Yes. When you have to keep fighting against the tool it’s better to give up. They have spent millions making it exactly like that my little prompt won’t fix it.

Re: Claudette: Make Claude stop talking like a BuzzFeed article

#227

Post by jobuildsstuff » Sat, Aug 22, 2026, 4:32 PM UTC

[flagged]

Re: Claudette: Make Claude stop talking like a BuzzFeed article

#228

Post by SquireBuilds » Sat, Aug 22, 2026, 4:34 PM UTC

Following a version of this by trial and error and using the usual 'no em-dashes' and don't explain/comment unless there's a non-obviuous reason.

Re: Claudette: Make Claude stop talking like a BuzzFeed article

#229

Post by nrmitchi » Sat, Aug 22, 2026, 4:52 PM UTC

lucisferre wrote: post #225

nrmitchi wrote: post #84

Earlier quoted context omitted.

I do not have evidence or data that supports this. It is only my thought. Claude, since Opus 5, speaks more and more like a wannabe-thought-leader pontificating on social media for engagement. Everything is a bait-then-switch, or a multi-post story format. The "engagement" that works well for social media makes actual work extremely frustrating. My unsupported belief is that this is caused by an obnoxious number of p…

Evidence or not, this is probably the most satisfying explanation I've heard for this behaviour. My own suspicion is that LLMs are not becoming more general-purpose over time, as these companies had hoped, and ongoing development in that direction is stalling. They will likely need to move in the direction of more specialization and train LLMs for specific use-cases. However, this would also be admitting that they ar…

To be clear I do not believe it is intentional (nor do I believe I am somehow smarter or more knowledgable than the people running these processes).

I believe it is (very) unfortunately converging on the communication style that currently makes the most money as far as publicly available communication goes. Unfortunate, but not entirely unexpected.

Re: Claudette: Make Claude stop talking like a BuzzFeed article

#230

Post by bonoboTP » Sat, Aug 22, 2026, 4:56 PM UTC

unclebucknasty wrote: post #221

bonoboTP wrote: post #216

Earlier quoted context omitted.

It's an empty point and a boring one and leads nowhere. It means nothing, but sounds deep. It's so shallow, any taxi driver and sociologist can come up with it and has, already in December 2022.

> It's an empty point and a boring one and leads nowhere. But, is it true ? > It means nothing...shallow... Ironically, the entirety of your comment just repeats that the GP comment means nothing. There is no further explanation or "depth".

Yours added no depth either.