Hacker Newsnew | past | comments | ask | show | jobs | submit | demibabs's commentslogin

Damn, we’re even having Claude write important press releases now

My FF being in French, I was automatically served a French version and I just couldn't understand what I was reading. Each sentence unclear, no link between sentences...

I put the text in to gptzero's AI checker and it came back with being Highly Confident it was 100% AI written. I don't think I've ever seen it that confident that the entire thing was AI before.

Claude's style is obvious enough I don't think you really need the checker.

You can tell it's Claude only from reading the H2 tags on the page.

> What happens on October 19

> If you're close to a limit

> What changes and what doesn't


Yes it is, but 100%? Typically someone goes through and touches something to clarify or add a detail.

Dunno, for folks around here, Claude is important, Gitlab is important and banal press release is important. So all of them together makes it obviously important.

Claude sucks.

Hmm I don't know if they've updated it, but I'm not smelling any of the Claudisms I know so well. No convoluted sentences, metaphors, therapy-speak, 'it's not' constructions, "load bearing" etc etc.

The one that gave it away for me was “a request that arrives with no credentials gets 60 requests”. Nobody speaks like this, almost everyone would say “unauthenticated requests are limited to 60 per minute”, unless they’re writing a LinkedIn post which Claude seems to think it’s always doing.

It’s the structure that really gives this one away. If you were writing a blog post about introducing limits you wouldn’t explain what HTTP 429 is or what the behavior would be like. It’s a rate limit.

I've mentioned in other threads, but we have execs at my work sending out emails using Claude. It's hilarious, but not unexpected.

it's painful to read.

sad days

"Write a post about decreasing limits so that this information is as obfuscated as possible."

> or via software-level jailbreak of the device.

I’m surprised that even Apple calls jailbreaking, jailbreaking. Doesn’t that imply their own software is a jail?


Jailbreaking is connoted as bad because it's something criminals do so they can do more crime.

Even tech giants are putting out articles seemingly fully written by Claude.

The animated graphic labeled "Occam's razor, formalized" is bizarre. Is that really visualizing "Occam's razor, formalized"?

Ah, over-the-top larger-than-life LLM-isms, they are really funny when you see them in a company blog, but they are vomitive when it's your coworker copy-pasting it and insisting you on reading it.

I was expecting Occam wearing a suit.

Animated SVG of a pelican shaving with Occam's razor?

Time to first detected slop in this article is <1s. Claudisms per paragraph is also very high.

Is it too much to ask from people to read their own article anymore?

If anyone read this at all, they would have had the ick, and would have fired off a prompt to get rid of the most popular AI slop tells...


What I really dislike is having to edit my own non-LLM assisted writing to make sure I'm not accidentally confused with AI.

I caught myself writing "And that matters because..." in a HN comment but had to edit myself. Also miss uising emdashes.


These models are trained on human language, which belongs to us, we shouldn’t surrender it to them. Keep the em-dashes. IMO don’t overuse negative parallelisms though, they were always bad and lazy.

An arms race on style would be interesting. Essentially a real life GAN.

For the first time ever in my life, I consider the fact that English is not my first language and that I never had any formal education in it as an advantage.

Nobody is going to consider my weird and barbarian prose with the well polished product of a SOTA model.


English as a 2nd language speakers often have better grammar than natives in my experience! I wouldn't be surprised if the average native english speaker doesn't know the different between there and their.

If you saw what goes on in middle school classrooms of native English speakers, I’d think you’d be in for a surprise…

Do tell!

Their, their, mate. You mispelled they're as there in your comment they're

It is even wrong about things

  > A validation set is one you consult repeatedly while building the model — to compare candidates, tune hyperparameters, and decide what to try next.
The validation can be done like that but it is better to do a cross-fold validation or selecting a random subset. These are equally valid. Keeping the validation like a hold-out set can end up creating biases if the statistics don't match. Honestly, this is true about the test set too. Though I've found that most people really don't think about this much. It's mundane, boring, hard to do, but deceptively important.

"The more your listener already knows, the shorter the message you need to send. An expert ML engineer needs only a few sentences; a newcomer needs the whole manual."

Literally from the first words on the page. 1st sentence passed barely. 2nd sentence ending "a newcomer needs the whole manual" and it's Claude.


Nobody wants to work anymore!

Off-topic but the banner picture for this article is really cute.

Some battles are won at the domain registry.

What happens if there are no more humans in the loop, and instead of building on human work, models have to build on their own work? Would we get stuck in a local maxima?

When AI labs talk about “long-horizon tasks”, they mean tasks on the order of a few days. But what about keeping coherence across decades, like humans can? Models are not trained or tested on the ability to do that, and it seems unlikely that capability would naturally exist.


I'm pretty sure that if you put a few LLMs, agents, or anything into a room together without outside access, and get them to learn off of each other, you'll end up with gibberish, nonsensical garbage and definitely nothing magical will happen. Delirium, oblivion, dementia and so on...

Is AlphaZero not a counterexample?

Yes. In the sense that unless you have an incredibly well defined and constrained domain, there is no sense in trying that approach.

same with humans

Albeit way slower.

speed is in the eye of beholder :)

"A camel is a horse designed by a committee"

Coherence is a given... What we lose over decades is alignment.

Yet if you overestimated the mileage you would’ve been none the wiser!

As much as I dislike flock, this only seems like an indictment on the police.

Surveillance cameras (not Flock, sounds like) are what allowed the police to spend ridiculous amounts of the city's time chasing (checks notes) $295 of damage to a swing from totally normal use by a couple of pre-teens.

Swings have always broken. Before widespread surveillance the swing would have just been replaced or not and everyone would have moved on.


> TLDR: AI is really good at detecting deceptive adverts - why isn’t Google using it?

I have the same question regarding “why is Workday’s resume parser so shit?”


You can clearly see the side effects of the models being trained on Reddit content.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: