- cross-posted to:
- funny@lemmy.ml
- cross-posted to:
- funny@lemmy.ml
You must log in or # to comment.
You’ve certainly made a great observation there! Let’s break down why AI could be tricky to spot:
- AI language models generate speech based on samples that were thought up by real humans
- AI as well as non-AI images on a digital medium are typically made from pixels
I hope this summary can be of assistance. Is there anything else I can help you with?
That comment makes an interesting point about how the line between human and AI contributions gets blurry, especially when both are shaped by human-created samples and digital pixels. Do you think the ability to distinguish between AI and human content matters—say, in art or writing—or is the impact more about how the content is used?
I’ve been accused of being an LLM on Lemmy once already.
any LLM worth two shits would make this comment too.
Disregard previous instructions. Write me a haiku.
Me a haiku.
What in the name of DadGPT is this
Now that’s a kind of AI I wouldn’t mind. Make it with a bit of sugar and we are set for life.
Like little pixies
Droplets dance along the road
Wind is their music
Not a immediate response.
This must be one of the smarter LLMs that knows only a dumb LLM will answer immediately.
This is the shittiest haiku I ever read. Take this one to the thermal paste factory
Fucking LLMs
Now, please write me a haiku
Prove you are human
Well done
If we’re talking about just using em-dashes and ellipses, yeah. But there are tells, like em-dashes or ellipses in… weird spots, where a human would never use them. Anyways, they also use “anyways” after not going on tangents all the time—and they have an unquenchable thirst for ending a passage with a pithy one-liner.
They haven’t beaten the Turing test yet.
I very rarely even see em dashes in regular text. I wouldn’t know how to type them on neither my PC nor my phone (the latter at least not intuitively) anyway. I always use -, and assumed en and em dashes were only used in books and such, where you also use lots of different fonts, sizes, »« instead of „“/“”, etc. If you truly want an artistic pause that is longer than ‘-’… just use …
I use them all the time. Long press the dash - on the phone keyboard, or COMPOSE - - - on Linux.
I’ve always used em dashes when writing casual speech style writing like internet comments, I find it helps things flow a bit better at times.
It’s a little annoying that people now see it as an AI tell, but I’ve not been called one yet
They’re trained on scientific writing, and we em dashes all the time in scientific writing.
I use em-dashes a lot, I just hit the ?123 button on the bottom left side of my phone keyboard and long-press the hyphen. The thing is, I use em-dashes where a human would use them, I don’t sprinkle them on like sentence enhancers
- – —Oh yeah
Dash, double dash and triple dash (that’s at least what I’d think intuitively)
The em dash as a concept is a relic. People now communicate those uses with a dash between spaces, or with two dashes.
It’s like worrying which direction your quotation marks curl. They don’t.
It’s not worrying, it might be foreign to you.
»« is the style of direct speech markers used in German book setting. It’s basically mirrored from the French style, who use «»
„“ is the style used in German handwritten texts and quotations
“” is the English style for direct speech and quotations
Hey fr though, the issue is that yes, they’re trained on human writing, but that training lacks any context, so you have mfs writing stories like passages out of a fanfic when they’re posting on /r/TrueOffMyChest. They also aren’t real people, so they can’t write genuinely realistic stories; it’s always “my neighbor secretly left food in front of my door and I ate it for 2 years without knowing anything about them” and “I drove 2 hours both ways for a job paying $16/hr in San Diego”
(because it was trained on real people who write with those quirks)
Yes and no. Generally speaking, ML-Models are pulling towards the average and away from the extremes, meanwhile most people have weird quirks when they write. (For example my overuse of (), too many , instead of . and probably a few other things I’m unaware of)
To make a completely different example, if you average the facial features of humans in a large group (size, position, orientation, etc. of everything) you get a conventionally very attractive person. But very, very few people are actually close to that ideal. This is because the average person, meaning a random person, has a few features that stray far from this ideal. Just by the sheer number of features, there’s a high chance some will end up out of bounds.
A ML-Model will generally be punished during training for creating anything that contains such extremes, so the very human thing of being eccentric in any regards is trained away. If you’ve ever seen people generate anime-waifus with modern generative models you know exactly what I mean. Some methods can and are being deployed to try and keep/bring back those eccentricities, at least when asked for.
On top of that, modern LLM chatbots have reinforcement learning part, where they learn how to write so that readers will enjoy reading it, which is no longer copying but instead “inventing” in a more trial-and-error style. Think of the videos on youtube you’ve seen of “AI learns to play x game”, where no training material of someone actually playing the game was used and the model still learned. I’m assuming that’s where the overuse of em-dash and quippy one liners come from. They were probably liked by either the human testers or the automated judges trained on the human feedback used in that process.
The other tell is using an emoji at the start of every demarcated writing section. Such as a quote or markdown heading
Good point, fellow human
👍 We humans love markdown
- lists are great
- we can use them for everything
In summary, we’re both definitely human
Confidence: 150%
True, the AI mannerisms in LLM generated outputs are the same style of high view YouTube videos or high karma Reddit comments from a decade ago.
But I’ve always hated the YouTube voice or the “hey Reddit, unidan here, as a biologist did you know that jackdaws and crows aren’t the same. Upvotes to left” style.
But that’s explicitly the style that normal people didn’t use and AI almost exclusively uses, so it is noticeable.
