I’m gonna use it manually but in the wrong — places.
One of my colleagues submitted a PR with a bunch of emojis in the readmes and log statements and I’m just so infuriated with it.
Our leadership made using and excelling at Copilot as one of our Key Results for upskilling, I make it point to make my code look as botty as possible to show how serious I’m about achieving the target 🎯
I just made a CI pass to forbid non ASCII characters in the code. Found a lot of em dashes :(
There are plenty of non ASCII characters that are okay in code. ñ comes to mind. There are also box drawing characters.
We don’t use them in my project, I only added an exception for ©®™ and such. You can easily whitelist any character range you need. My command looks like this:
- (! grep -r -I -P '[^\x{00}-\x{7f}©®™°]' src)This is goofy, I’m not gonna fail a build because somebody used some random Unicode character. That’s draconian.
I literally see it everywhere in my companies’ documentation
In the logs???
Yeah, they used ⚠️ in a warning.
They add excessive logging too. I had cursor write some UI code and I thought my console was going to explode. Console logs each step of the way haha. I guess that’s the easiest way for it to debug its own code
AGI is science fiction and never happening
I wouldn’t say never, but certainly not this century. Probably not the next one, either.
And at this point we might as well refer to it as something like “Machine Consciousness”, because “AI” an any derivatives of it have been rendered garbage for actually talking about the sci-fi concept.
AI was never used to refer to human like intelligence anyway in the AI field. Like any form of computer “intelligence” was called AI. Like chess machines or NPC logic in a game were called AI. It just needs to seem intelligent like the word artificial implies. It’s only now with these LLM peddling companies that suddenly the general public thinks AI refers to reaching human intelligence.
AI has been used to refer to human like intelligence in fiction for at least half a century.
OpenAI in 2020: Proves its own models will never, ever, reach 95% human accuracy.
https://arxiv.org/pdf/2001.08361
DeepMind: Corrects their math, limit is still <95% with Infinite power and training.
https://arxiv.org/pdf/2102.06701
Investors: “SHUT UP AND TAKE MY MONEY!”
Totally not a bubble! Just a few thousand more data centers!
Bad news, everyone: your AI detection skills are now useless. Every comment here including mine is AI slop.
Dude, Ive been writing for a couple of years now. And I use em dashes, probably because all the authors I like use them. As soon as this AI thing started doing it, everything Ive ever written turned into AI slop over night lol.
Ok, what the heck is an em-dash?
There are two types of dashes. One is the “n-dash” (or “en-dash”), which takes up one space, and is most often used to hyphenate words; and the other is the “m-dash” (or "em-dash) which takes up two spaces, and is most often used to bracket off parenthetical information within a sentence, like kind of a lighter weight parentheses. Em-dashes get used a lot in novels and other published writing that is subject to correction from a professional copy editor, but very rarely in the daily typing of regular people. So now when people see it getting used they just assume it must be a clanker.
A slight correction, en-dashes are used mostly to indicate ranges like Mon–Fri. Hyphens are a separate third thing, smaller than an en-dash.
- hyphen
– en-dash
— em-dashThey get their names originally from having the same width as the letter n or m respectively in typesetting (though not all fonts follow that necessarily).
Thank you for this!!!
Thank you for the synopsis. However;
This is a clanker:
and this is software:
M-dashes are super useful for setting off parentheticals nicely.
You should be able to figure it out—even if you don’t know what you’re looking for—if you’re sent the proper response ;)
Didn’t even use em dashes lol
Lmao fat-fingered my keeb when typing. Edited my original comment. I should have coffee before posting first thing in the morning :)
What is a search engine?
It’s the tell that someone used an LLM to write the response.
That’s absurd. I’ve long used em dashes in writing. It’s not at all a tell. Plenty of people use them.
Plenty = dozen.
Most just use n-dashes. M-dash doesn’t even exist on most virtual keyboards.
A lot of software will autocorrect two hyphens into an em-dash, but it’s not consistent everywhere.
This is how I’ve used them. Or I’ll just put – as a place holder and use the replace all function to change them when I finish. Luckily I’ve never had anyone accuse me of AI writing but it’s probably a matter of time. I think the use of em-dash as the “telltale sign” of AI is silly. They’re just more often used in research, documentation, and academics which the AI is heavily trained on. Obviously when average reddit Joe uses them frequently it’s a bit of a red flag, so the context of the writing is key here. The AI uses it because it is widely and commonly used by humans.
Ok, but what is it and how is it different than a dash?
I was checking out the online reviews of my dad’s gastroenterologist the other day. The reviews all contained em-dashes although it was obvious just from the content that they were AI-generated. Dude is full of shit, ironically enough.
Just a few billion more…? 🤲
I heard Altman wants a trillion $ bailout
A development this groundbreaking is definitely worth another 12-figure investment.
I’m just mad I can’t use dashes as punctuation anymore
*Without it looking like I’m using chatgpt
Same. Same.
Now you can’t write without using em-dashes or you’ll be called a bot.
Looking pretty sus there using a dash in emdash
So is there a sea horse emoji or not?
🦘 yes
Close enough 🤣
wait, why isn’t this a thing tho?
Because then we would need a pregnant male seahorse emoji and the world isn’t ready for that.
That makes me want it more.
But we already have pregnant male humans…
🫃🫄
By now, I feel they are just straight trying to appeal to scammers.
Open ai is a joke.
Fuck that stupid AI/AGI narrative, even if this is meant as a joke. They’re LLMs. Some of them not even so large. A bunch of if statements with access to all the data that advertising companies like Google have harvested for decades.
You’re thinking of expert systems, an old form of AI from the 70s
Wonder if you specify to use en dashes instead if it just collapses and the simulation resets
Yay. Next step… AGI.
AGI is a scam.
