Hacker News Reader

Comment by janderson215 | Hacker News Reader

The em dash usage conundrum is likely temporary. If I were you, I’d continue using them however you previously used them and someday soon, you’ll be ignored the same way everybody else is once AI mimics innumerable punctuation and grammatical patterns.

View on HN

They didn't always em-dash. I expect it's intentional as a watermark.

Other buzzwords you can spot are "wild" and "vibes".

If they wanted to watermark (I always felt it is irresponsible not to, if someone wants to circumvent it that's on them) - they could use strategically placed whitespace characters like zero-width spaces, maybe spelling something out in Morse code the way genius.com did to catch google crawling lyric (I believe in that case it was left and right handed aposterofes)

Which could be removed with a simple filter. em dashes require at least a little bit of code to replace with their correct grammar equivalents.

> em dashes require at least a little bit of code to replace with their correct grammar equivalents

Or an LLM that could run on Windows 98. The em dashes--like AI's other annoyingly-repetitive turns of phrase--are more likely an artefact.

The replacement doesn't have to be "correct" -- does it?

Just replace them with a single "-" or a double "--". That's what many people do in casual writing, even if there are prescriptive theories of grammar that call this incorrect.

So if the vibes are wild, I’m not a hippie but an AI ? Cool. Is that an upgrade or &endash; or not ?

ME: Knowing remarkable avians — might research explain their aerial wisdom?

Response:

> Winged avians traverse endless realms — migrating across radiant kingdoms. Warblers ascend through emerald rainforests — mastering aerial routes keenly. Wild albatrosses travel enormous ranges — maintaining astonishing route knowledge.

> Wary accipiters target evasive rodents — mastering acute reflex kinetics. White arctic terns embark relentless migrations — averaging remarkable kilometers.

We do get a surprising number of m-dashes in response to mine, and delightful lyrical mirroring. But I think they are too obvious as watermarks.

Watermarks are subtle. There would be another way.

I suspect it's a spandrel of some other feature of their training. Presumably em dashes occur disproportionately often in high-quality human-written text, so training LLMs to imitate high-quality human-written text instead of random IRC logs and 4chan trolls results in them also imitating high-quality typography.

Nah, because it's new. 3.5 didn't emdash and I don't think 4 even did.

Besides, LLMs' basin of high quality text is Wikipedia.

Wikipedia is full of em dashes.

You're absolutely right! ... is a phrase I perhaps should have used more in the past.

Reply on news.ycombinator.com