All this really does is show areas where the writing requirements are already bullshit and should be fixed.
Like, consumer financial complaints. People feel they have to use LLMs because when they write in using plain language they feel they're ignored, and they're probably right. It suggests that these financial companies are under regulated and overly powerful. If they weren't, they wouldn't be able to ignore complaints when they're not written in lawyerly language.
Press releases: we already know they're bullshit. No surprise that now they're using LLMs to generate them. These shouldn't exist at all. If you have something to say, don't say it in a stilted press-release way. Don't invent quotes from the CEO. If something is genuinely good and exciting news, make a blog post about it by someone who actually understands it and can communicate their excitement.
Job postings. Another bullshit piece of writing. An honest job posting would probably be something like: "Our sysadmin needs help because he's overworked, he says some of the key skills he'd need in a helper are X, Y and Z. But, even if you don't have those skills, you might be useful in other ways. It's a stressful job, and it doesn't pay that well, but it's steady work. Please don't apply if you're fresh out of school and don't have any hands-on experience." Instead, job postings have evolved into some weird cargo-culted style of writing involving stupid phrases like "the ideal candidate will..." and lies about something being a "fast paced environment" rather than simply "disorganized and stressful". You already basically need a "secret decoder ring" to understand a job posting, so yeah, why not just feed a realistic job posting to an LLM and make it come up with some bullshit.
Job postings are wild. Like, "Java Spring Boot developer with 8+ years experience" would be fine 90% of the time.
Exactly. LLM's assisting people in writing soul-sucking corporate drivel is a good thing, I hope this changes the public perception on the umbrella of 'formal office writing'. (including: internal emails, job applications etc.) So much time-wasting bullshit to form nothing productive.
As a person who is intrigued in linguistics, I wonder how
AILLMs will affect real languages. I wonder if there is any research papers on this.Not quite the same, but I'm waiting for the day when people will pronounce street names like the GPS, instead of how they are actually pronounced. The street Schoenherr, in my neck of the woods is pronounced "Shane urr (yes, like the planet Omicron Percii 8, cause Detroit (Day twah) is weird), but the GPS says "Shown her". I'm really curious to see how long it takes for the computer voice to be considered the correct one.
I dunno. If people can’t be bothered to write stuff anymore, I doubt they will be bothered to read it either. Also, the model deviates towards the mean by its very design.
Paging @[email protected] :)
I am not saying the two are equally comparable, but I wonder if the same "most rapid change in human written communication" could also have been said with the proliferation of computer-based word processors equipped with spelling and grammar checks.
Llm detectors are always snake oil 100% of the time. Anyone claiming otherwise is lying for personal gain.
Source?
They make the claim, the burden of proof is on them. Please look at the paper, there is so much hand waving it could be a parade.
The source is their fingers when they typed in the message. Silly goose.
I just want to point out that there were text generators before ChatGPT, and they were ruining the internet for years.
Just like there are bots on social media, pushing a narrative, humans are being alienated from every aspect of modern society.
What is a society for, when you can't be a part of it?
I just want to point out that there were text generators before ChatGPT, and they were ruining the internet for years.
Hey now, King James Programming was pretty funny.
For those unfamiliar, King James Programming is a Markov chain trained on the King James Bible and the Structure and Interpretation of Computer Programs, with quotes posted at https://kingjamesprogramming.tumblr.com/
4:24 For the LORD will work for each type of data it is applied to.
In APL all data are represented as arrays, and there shall they see the Son of man, in whose sight I brought them out
3:23 And these three men, Noah, Daniel, and Job were in it, and all the abominations that be done in (log n) steps.
I was first introduced to it when I started reading UNSONG.
This was such a good idea, so many of these are fire.
then shall they call upon me, but I will not cause any information to be accumulated on the stack.
How much more are ye better than the ordered-list representation
evaluating the operator might modify env, which will be the hope of unjust men
I feel like the term "touch grass" applies to this comment more than anything.
Speak for yourself
“Touch grass” is exclusionary and divisive. I asked Copilot to explain why.
The expression "touch grass," often used to suggest someone take a break and reconnect with nature, can unintentionally be exclusionary, especially for people living in arid regions where grass is not readily available. For instance, in parts of the Middle East, where the landscape is predominantly desert, grass is a scarce resource and the idea of "touching grass" might feel disconnected from reality. This phrase overlooks the diverse ways people around the world engage with their environment and can unintentionally alienate those who don't have easy access to grassy areas. It’s a reminder that language should be mindful of diverse experiences and environments.
I'm the type to be in favor of new tech but this really is a downgrade after seeing it available for a few years. Midterms hit my classes this week and I'll be grading them next week. I'm already seeing people try to pass off GPT as their own, but the quality of answers has really dropped in the past year.
Just this last week, I was grading a quiz on persuasion and for fun, I have students pick an advertisement to analyze. You know, to personalize the experience, this was after the super bowl so we're swimming in examples. Can even be audio, like a podcast ad, or a fucking bus bench or literally anything else.
60% of them used the Nike Just Do It campaign, not even a specific commercial. I knew something was amiss, so I asked GPT what example it would probably use it asked. Sure enough, Nike Just Do It.
Why even cheat on that? The universe has a billion ad examples. You could even feed GPT one and have it analyze for you. It'd be wrong, cause you have to reference the book, but at least it'd not be at blatant.
I didn't unilaterally give them 0s but they usually got it wrong anyway so I didn't really have to. I did warn them that using that on the midterm in this way will likely get them in trouble though, as it is against the rules. I don't even care that much because again, it's usually worse quality anyway but I have to grade this stuff, I don't want suffer like a sci-fi magazine getting thousands of LLM submissions trying to win prizes.
Why even cheat on that? The universe has a billion ad examples.
I'm not one of your students, but I do remember how I thought in high school. Both of my parents worked, so I was the one that had to cook dinner and help my little brothers with their homework, then I had multiple hours of my own homework to do.
While I do enjoy analyzing media, the homework I struggled with would get priority. I was the oldest, so I didn't have anybody to ask for help with questions, and often had to spend a larger amount of time than intended on topics I struggle with. So, I'd waste the whole night struggling with algebra and chemistry, then do the remaining 'easy' assignments as quickly and carelessly as possible so I could get to bed before midnight. Getting points knocked off for shoddy work is far preferable to getting a zero for not doing it at all, and if I could get to bed at a reasonable time, I wouldn't lose points in the morning class for falling asleep.
It just... makes sense to cheat sometimes.
Students and cheating is always going to be a thing, only the technology evolves. It's always been an interesting cat and mouse game imo, as long as you're not too personally affected (sorry).
I was a student when the internet started to spread and some students had internet at home, while most teachers were still oblivious. There was a french book report due and 4 kids had picked the same book because they had found a good summary online. 3 of the kids hand wrote a summary of the summary, 1 kid printed out the original summary and handed that in. 3 kids received a 0, the 4th got a warning to not let others copy his work :D
Lol, well sounds like a bad assignment if you can get away with just summary, although I guess it is language class(?) it's more reasonable. I'm not really shooken up over this type of thing, though. I'm not pro-cheating, but it's not for justice or morality; it's cause education is for the students benefit and they're missing out on growth. We really need more critical thinkers in this world. Like, desperately need them. Lol
The reason chatgpt would recommend Nike though is because of its human-based training data. This means that for most humans the Nike ad campaign would also be the first suggestion to come to mind.
I'm not saying LLMs aren't having an impact, or denying that said impact is negative, but the way people talk about them is infuriating because it just displays a lack of understanding or forethought on how these systems work.
People always talk about how they can tell something "sounds like chatgpt" or, as is the case here, is the default chatgpt answer, while ignoring the only reason it would be so is because of the real human patterns which it is mimicking.
Brief caveats: of course chatgpt is wildly fallible and when producing purely generative content it pulls from nowhere because it's just remixing unrelated sources, but for things within the normal course of discussion and output chatgpt's output is vastly more human-like than we want to pretend.
I would almost guarantee that Nike's "just so it" was the singularly most popular answer to this kind of assignment before chatgpt existed too.
Except I've given this quiz prior to GPT and no, it wasn't once used because it's not even a current advertisement campaign. My average 19 year old usually uses examples from my influencers, for instance, so I get stuff like Hello Fresh or Better Help, and usually specific to an ad read on stream on the past couple weeks. After all, the question asks for ads they've seen and remembered.
Also, you neglect how these models get data. It's likely pulled not because it's a favorite, but because GPT steals from textbooks, blogs, etc, and those examples that would use that as a go-to (especially if the author uses 90s examples). Plus nevermind that your joe shmo Internet user isn't the same as the group I'm teaching, most of them weren't even alive when the Just Do It campaign started, lol.
It really undermines the point of coming up with your own examples and applying theory to something from their life. I am not inherently anti GPT but this is a very bad use case.
As someone who has been a teenager. Cheating is easy, and class wasn't as fun as video games. Plus, what teenager understands the importance of an assignment? Of the skill it is supposed to make them practice?
That said, I unlearned to copy summaries when I heard I had to talk about the books I "read" as part of the final exams in high school. The examinor would ask very specific plot questions often not included in online summaries people posted... unless those summaries were too long to read. We had no other option but to take it seriously.
As long as there isn't something that GPT can't do the work for, they won't learn how to write/do the assignment.
Perhaps use GPT to fail assignments? If GPT comes up with the same subject and writing style/quality, subract points/give 0s.
Last November, I gave some volunteer drawing classes at a school. Since I had limited space, I had to pick and choose a small number of 9-10yo kids, and asked the students interested to do a drawing and answer "Why would you like to participate in the drawing classes?"
One of the kids used chatgpt or some other AI. One of the parts that gave it away was that, while everyone else wrote something like "I want because", he went on with "By participating, you can learn new things and make friends". I called him out in private and he tried to bullshit me, but it wasn't hard to make him contradict himself or admit to "using help". I then told him that it was blatantly obvious that he used AI to answer for him and what really annoyed me wasn't so much the fact he used it, but that he managed to write all of that without reading, and thought that I would be too dumb or lazy to bother reading or to notice any problems.
I have a similar background and no surprise, it's mostly a problem in my asynchronous class. The ones who have my in person lectures are much more engaged, since it is a fun topic and I don't enjoy teaching unless I'm also making them laugh. No dice with asynchronous.
And yeah, I'm also kinda doing that with my essay questions, requiring stuff you sorta can't just summarize. Important you critical thinking, even if you're not just trying to detect GPT.
I remember reading that GPT isn't really foolproof on verifying bad usage, and I am not willing to fail anyone over it unless I had to. False positives and all that. Hell, I just used GPT as a sounding board for a few new questions I'm writing, and it's advice wasn't bad. There's good ways to use it, just... you know, not so stupidly.
"I recall Ethan Mollick discussing a professor who required students to use LLMs for their assignments. However, the trade-off was that accuracy and grammar had to be flawless, or their grades would suffer. This approach makes me think—we need to reshape our academic standards to align with the capabilities of LLMs, ensuring that we’re assessing skills that truly matter in an AI-enhanced world.
That's actually something that was discussed like, two years ago within the institutions I'm connected to. I don't think it was ever fully resolved, but I get the sense that the inaccurate results made it too troublesome.
My mentally coming out of an education degree, if your assessment can be done by AI, you're relying too much on memorization and not enough on critical thinking. I complain in my reply, but the honest truth is these students mostly lost points because they didn't apply theory to the example (although it's because the example wasn't fully understood since it wasn't their own). K-12 generally fails on this, which is why freshmen have the hardest time with these things, GPT or otherwise.
I was just commenting on how shit the Internet has become as a direct result of LLMs. Case in point - I wanted to look at how to set up a router table so I could do some woodworking. The first result started out halfway decent, but the second section switched abruptly to something about routers having wifi and Ethernet ports - confusing network routers with the power tool. Any human/editor would catch that mistake, but here it is.
I can only see this get worse.
The Internet was shit before LLMs
It had its fair share of shit and that gradually increased with time, but LLMs are like a whole new level of flooding everything with zero effort
I'd say it was weird, not shit. It was hard to find niche sites, but once you did they tended to be super deep into the hobby, sport, movies, or games.
SEO (search engine optimization) was probably the first step down this path, where people would put white text on a white background with hundreds of words that they hoped a search engine would index.
It's not just the internet.
Professionals (using the term loosely) are using LLMs to draft emails and reports, and then other professionals (?) are using LLMs to summarise those emails and reports.
I genuinely believe that the general effectiveness of written communication has regressed.
Yep. My work has pushed AI shit massively. Something like 53% of staff are using it. They're using it to write reports for them for clients, all sorts. It's honestly mad.
How did they estimate whether an LLM was used to write the text or not? Did they do it by hand, or using a detector?
Since detectors are notorious for picking up ESL writers, or professionally written text as AI-Generated.
They developed their own detector described in another paper. Basically, this reverse-engineers texts based on their vocabulary to provide an estimate on how much of them were ChatGPT.
This sounds plausible to me, as specific models (or even specific families) do tend to have the same vocabulary/phrase biases and “quirks.” There are even some community “slop filters” used for sampling specific models, filled with phrases they’re known to overuse through experience, with “shivers down her spine” being a meme for Anthropic IIRC.
It’s defeatable. But the “good” thing is most LLM writing is incredibly lazy, not meticulously crafted to avoid detection.
This is yet another advantage of self hosted LLMs as they:
-
Tend to have different quirks than closed models.
-
Have finetunes with the explicit purpose of removing their slop.
-
Can use exotic sampling that “bans” whatever list of phrases you specify (aka the LLM backtracks and redoes it when it runs into them, which is not normally a feature you get over APIs), or penalizes repeated phrases (aka DRY sampling, again not a standard feature).
-
They just asked a few people if they thought it was written by an LLM. /s
I mean, you can tell when something is written from ChatGPT, especially if the person isn't using it for editing, but is just asking it to write a complaint or request. It is likely they are only counting the most obvious, so the actual count is higher.
I don't know of any reason that the proportion of ESL writers would have started trending up in 2022.
If it's due to LLM is it "human written communication"?
Human-like written communication.
Even if it was fully AI generated its still human communication in a written format, at least until the AIs start writing to each other without a human intermediary.
I thought there was a social network that is completely filled with AI and no real humans.
Edit: found it https://socialai.co/
https://www.theverge.com/2024/9/17/24247253/social-ai-app-replace-humans-with-bots
Written communication of humans, not by humans
„signs of LLM writing” doesn’t mean that the whole thing was written by it.
This is the top result on duck duck go for how tall does a soursop tree get:
https://livetoplant.com/soursop-plant-size-get-the-right-size-for-you/
Gee thanks, I'm cured.
Btw does any one know if Soursops have an aggressive root system?
Well if your books start talking back you should get help. The computer just started getting good (I remember Dr Sbaitso)
That’s scary shit. Hopefully this can slow down some.
Who wants to be licked by an emo?
omg me
Don’t worry, I see you
BREAKING NEWS: Since the invention of calculators, less people using abacus!
Not a good analogy, except there is one interesting parallel. My students who overuse a calculator in stats tend to do fine on basic arithmetic but it does them a disservice when trying to do anything more elaborate. Granted, it should be able to follow PEDMAS but for whatever weird reason, it doesn't sometimes. And when there's a function that requires a sum and maybe multiple steps? Forget about it.
Similarly, GPT can make cliche copy writing, but good luck getting it to spit out anything complex. Trust me, I'm grading that drinble. So in that case, the analogy works.
What a dumb comparison. Calculators are just tools to do the same mechanical action as abaci, which were also just tools to speed up human mechanical actions of calculation.
Writing, drawing, research are creative, not mechanical, and offloading them to a tool is very different from offloading calculations to integrated circuits
This. It's a tool, embrace it and learn the limitations....or get left behind and become obsolete. You won't be able to keep up with people that do use it.
dude you figuring out how to make the AI shit out something half-passable isn't making you clever and superior, it's just sad
The invention of the torque wrench didn’t severely impede my ability to retrieve stored information, and everyone else’s, affecting me by proxy.
The tech four years ago was impressive but for me it’s only done two things since becoming widely available: thinned the soup of Internet fun things, and made some people, disproportionally executives at my work, abandon a solid third of their critical thinking skills.
I use AI models locally, to turn around little jokes for friends, you could say I’ve put more effort into machine learning tools than many daily AI users. And I’ll be the first to call the article described by OP as a true, shameful indictment of us as a species.
