@SnoopJ @glyph @jalefkowit I can not imagine an explanation of how this occurred that isn't damning.

@mttaggart @glyph @jalefkowit laying my personal wager on "someone used the nonsense machine" but agreed, I cannot think of any explanation for this that would be remotely acceptable.

@mttaggart @glyph @jalefkowit correction: I *can* think of one, and it's "they accidentally quoted from the hit-piece blog articles instead"

But they didn't do that, those don't contain the quotes either.

@SnoopJ@hachyderm.io @glyph@mastodon.social @jalefkowit@vmst.io @mttaggart@infosec.exchange I am disappointed, although not surprised, that my read on Benj's work at Ars turned out to the accurate one in recent years. Bleh.

@aud very rude of him to fall off so strongly immediately after I defended his work to you

@mttaggart

Seeing the absolute shit shoveling going on that counts for producing content at Ars over the past well...10-15 years at this point...

I'm not sure anyone should be following them let alone paying them subscription money.

That they are now using AI to write their articles is the least surprising thing ever. What a catastrophic fall from grace.

@SnoopJ@hachyderm.io One explanation I can think of is that an editor or Conde Nast higher up generated this using the author's bylines, without their involvement.

I would not put it past Conde Nast.

@SnoopJ@hachyderm.io but if his work had been good (which I was not familiar with before I started reading his articles that smacked increasingly more of boosterism) before... well, it certainly seems to fall into the pattern that people are seeing with regards to LLM usage (trusting it more and more without checking, etc).

@mttaggart Seems they took it down. Just wow.

@HitokiriEric @mhoye @arstechnica Maybe, but Pepperidge Farm remembers: https://mttaggart.neocities.org/ars-whoopsie

UPDATE: They pulled the story, but I had it up and had SingleFile in my browser, so: https://mttaggart.neocities.org/ars-whoopsie

@SnoopJ @mttaggart @glyph Looks like the article has been completely deleted.

Aurich said in the comments before it vanished that they were aware of the issue and looking into it. I assume the findings were pretty dire.

Page not found 404 | Ars Technica

(arstechnica.com)

@mttaggart archive dot org link since they took it down: https://web.archive.org/web/20260213194851/https://arstechnica.com/ai/2026/02/after-a-routine-code-rejection-an-ai-agent-published-a-hit-piece-on-someone-by-name/

@mttaggart Crap, they DID?! I still have it open! Anything you want before it goes into the void?

@hackillu https://web.archive.org/web/20260213194851/https://arstechnica.com/ai/2026/02/after-a-routine-code-rejection-an-ai-agent-published-a-hit-piece-on-someone-by-name/

@mttaggart a "we have hallucinated too and have pulled the article that was here before" would have been better. Good thing you saved it.

This scoop brought to you by the TTI Intel Feed, which also routinely beats commercial threat intel to the punch on important emerging threats.

· The Taggart Institute Intel Center

(intel.taggartinstitute.org)

@mttaggart And thus we close it.

@mttaggart I got the comments though.

@hackillu Nice!