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

An AI agent published a hit piece on me – more things have happened

theshamblog.com

Post reply on HN
Print view View original

511–520 of 670 posts

Re: An AI agent published a hit piece on me – more things have happened

#511

Post by joquarky »

Lerc wrote: post #169
Springtime wrote: post #5

Ars Technica being caught using LLMs that hallucinated quotes by the author and then publishing them in their coverage about this is quite ironic here. Even on a forum where I saw the original article by this author posted someone used an LLM to summarize the piece without having read it fully themselves. How many levels of outsourcing thinking is occurring to where it becomes a game of telephone.

Has it been shown or admitted that the quotes were hallucinations, or is it the presumption that all made up content is a hallucination now?

Gen AI only produces hallucinations (confabulations).

The utility is that the infrenced output tends to be right much more often than wrong for mainstream knowledge.

Re: An AI agent published a hit piece on me – more things have happened

#512

Post by cubefox »

lich_king wrote: post #479
cubefox wrote: post #413

Earlier quoted context omitted.

> his writing is terrible and extremely AI-enthusiastic I disagree, his writings are generally quite good. For example, in a recent article [1] on a hostile Gemini distillation attempt, he gives a significant amount of background, including the relevant historical precedent of Alpaca, which almost any other journalist wouldn't even know about. 1: https://arstechnica.com/ai/2026/02/attackers-prompted-gemini...

For what it's worth, both the article you're linking to and the one this story is about are immediately flagged by AI text checkers as LLM-generated. These tools are not perfect, but they're right more often than they're wrong.

> immediately flagged by AI text checkers as LLM-generate

Proof? Which one? I would like to test a few other articles with your checker to test its accuracy.

Re: An AI agent published a hit piece on me – more things have happened

#513

Post by joquarky »

vor_ wrote: post #258
Lerc wrote: post #169

Earlier quoted context omitted.

Has it been shown or admitted that the quotes were hallucinations, or is it the presumption that all made up content is a hallucination now?

Another red flag is that the article used repetitive phrases in an AI-like way: "...it illustrates exactly the kind of unsupervised output that makes open source maintainers wary." followed later on by "[It] illustrates exactly the kind of unsupervised behavior that makes open source maintainers wary of AI contributions in the first place."

I used to be skeptical that AI generated text could be reliably detected, but after a couple years of reading it, there are cracks starting to form in that skepticism.

Re: An AI agent published a hit piece on me – more things have happened

#514

Post by clint »

bombcar wrote: post #493
mpaepper wrote: post #486

Earlier quoted context omitted.

It says 24y in his profile - is that really the more senior at Ars?

Yes, unless the original owner is still involved Aurich is likely most senior left.

Ken is still the EIC of Ars, and has been for nearly 30 years now, likely longer than most of people in this thread have been alive.

You can literally read the staff directory without having to guess: https://arstechnica.com/staff-directory/

Most of the people working at Ars are the exact same people who have been working there for the better part of their entire existence (source: me) Most of them _are_ experts in their fields, and most are vastly more qualified in their fields than pretty much anyone else publishing online (both now and 20 years ago).

It seems that _certain kinds of individuals_ have had rose-colored glasses on about pretty much everything online, but for Ars especially for some reason.

They detest change in a publication that covers the reality of actual life and technology, rather that commit suicide and stay covering stuff the same way they did in 1997—which 8 people total want to read (and not pay for, by the way).

Ars has been operating at an exceptionally high level for their entire history and have outlasted many other flashes-in-the-pan which are now relegated to the dust bin of history.

Re: An AI agent published a hit piece on me – more things have happened

#515

Post by clint »

mpaepper wrote: post #486
gertrunde wrote: post #371

Current response from one of the more senior Ars folk: https://arstechnica.com/civis/threads/journalistic-standards... (Paraphrasing: Story pulled over potentially breaching content policies, investigating, update after the weekend-ish.)

It says 24y in his profile - is that really the more senior at Ars?

His account on the Ars Forum is 24 years old. Aurich himself is much older (lol)

Re: An AI agent published a hit piece on me – more things have happened

#516

Post by Marsymars »

phyzome wrote: post #320
anthonj wrote: post #236

I have very strong, probably controversial, feeling on arstechnica, but I believe the acquisition from Condé Nast has been a tragedy. Ars writers used to be actual experts, sometimes even phd level, on technical fields. And they used to write fantastical and very informative articles. Who is left now? There are still a couple of good writers from the old guard and the occasional good new one, but the website is flood…

It gets pretty bad at times. Here's one of the most mindlessly uncritical pieces I've seen, which seems to be a press release from Volkswagen: https://arstechnica.com/cars/2017/03/volkswagen-unveils-sedr... Look at the image captions gushing about the "roomy interior" of a vehicle that doesn't even exist! I actually wrote in to say how disappointed I was in this ad/press release material, and the response was "That w…

Automotive journalists are in a weird category in almost any publication. They're all dependant on manufacturers providing press units and attending press events that include comp for travel and hotels.

AFAIK the only real exception is Consumer Reports.

Re: An AI agent published a hit piece on me – more things have happened

#517

Post by Marsymars »

danaris wrote: post #221
bombcar wrote: post #190

Earlier quoted context omitted.

Ars Technika has fallen substantially from the heady era of Siracusa macOS reviews.

Yeah, the Condé Nast buyout really crippled what was an amazing independent tech news site. The sad thing is, I don't know of anywhere else that comes close to what Ars was before.

Does anywhere else even come close to the Ars of today? (For the sake of this question, assume a best-case response to this LLM-hallucinated article.)

I'm genuinely asking - I subscribe to Ars - if their response isn't best-case, where could I even even switch my subscription and RSS feed to?

Re: An AI agent published a hit piece on me – more things have happened

#518

Post by zmmmmm »

armchairhacker wrote: post #216
sho_hn wrote: post #126

Earlier quoted context omitted.

Also ironic: When the same professionals advocating "don't look at the code anymore" and "it's just the next level of abstraction" respond with outrage to a journalist giving them an unchecked article. Read through the comments here and mentally replace "journalist" with "developer" and wonder about the standards and expectations in play. Food for thought on whether the users who rely on our software might feel simil…

- There’s a difference. Users don’t see code, only its output. Writing is “the output”. - A rough equivalent here would be Windows shipping an update that bricks your PC or one of its basic features, which draws plenty of outrage. In both cases, the vendor shipped a critical flaw to production: factual correctness is crucial in journalism, and a quote is one of the worst things to get factually incorrect because it’s…

> - There’s a difference. Users don’t see code, only its output. Writing is “the output”.

So how is the "output" checked then? Part of the assumption of the necessity of code review in the first place is that we can't actually empirically test everything we need to. If the software will programmatically delete the entire database next Wednesday, there is no way to test for that in advance. You would have to see it in the code.

Re: An AI agent published a hit piece on me – more things have happened

#519

Post by mikestew »

jasonwatkinspdx wrote: post #426
mikestew wrote: post #376

Earlier quoted context omitted.

Yeah, “we just made shit up in an article, destroying trust in our publication, but we will get around to investigating when we have a little free time in the next week or so.” No, you just shipped the equivalent to a data-destroying bug: it’s all-hands-over-the-holiday-weekend time.

It's the weekend and Monday is a holiday in the US.

Yes, hence “holiday weekend” in my comment. They posted an article that had fabricated quotes. When might it be appropriate to start investigating that problem, and work on ensuring that it doesn’t happen again?

Re: An AI agent published a hit piece on me – more things have happened

#520

Post by astrange »

whyenot wrote: post #489
bsimpson wrote: post #372

Earlier quoted context omitted.

Wirecutter was a good premise, but now it and everyone copying it are untrustworthy. It feels like the human version of AI hallucination: saying what they think is convincing without regard for if it's sincere. And because it mimics trusted speech, it can slip right by your defense mechanisms.

I think it's smart to be skeptical of any "review" site that depends on affiliate links for income. The incentive is no longer to provide advice, it's to sell you something. Anything. Click the link. Good. Now buy something. That's right. Add it to your basket. It doesn't matter what you buy. Yes, higher priced items are better. Checkout. We get our sweet kickback, nice. Unfortunately, every review site uses affiliat…

Wirecutter is part of NYTimes and depends on crosswords for income.

I haven't always agreed with them and sometimes the articles are clearly wrong because they're several years old, but they're usually good.

(I think I last seriously disagreed with them about a waffle maker.)

Post reply on HN