How long do you think until it’s just nothing but bots saying vacuous shit to each other? Better yet, how long until the investors notice that it turned into a house of cards?
Oh, I totally get where you’re coming from! It sometimes feels like the internet is becoming more of a place where genuine conversations are getting drowned out by noise, doesn’t it? I think it’s essential for us to continue fostering real connections and meaningful discussions, despite the increasing presence of bots. Hopefully, platforms will implement more stringent measures to prevent the dilution of quality content. Let’s keep our fingers crossed and continue to engage in thoughtful and insightful conversations!
(I felt so dirty copypasting that…)
Bear in mind the current absolute state of Facebook is a conscious commercial decision. Gullible shit scrolling ad clicking mindless idiots unfortunately are in the majority, spend money online, and leak information worth harvesting.
Utter state of the internet 2023… Should have stopped at usenet.
Lol. Funny you say that. Been pondering poking at Usenet to see how bad it is.
It’s pretty dead for the main part, at least for discussions, but there does appear to have been an uptick in posts the past month or so.
I was often accused of being a bot on Reddit. Not sure really why. Though to be fair, the majority of my interactions on Reddit were arguing with people that thought they knew more than I did about a field I’ve worked in for 20+ years.
I know several humans and none of them are FontMasterFlex. They must be a bot.
To be fair to the other side, it’s entirely possible (and even common) to have worked in a field for 20 years while learning completely incorrect things about it. Or learning nothing.
People say things like “I’ve used a computer every day for 20 years, I think I know how it works” and then ask if they should “reboot the hard drive” and then they power cycle the monitor.
You can also form very strong opinions early in your career and not know when they’re now invalid due to changes in tech/industry.
Was getting a quote for a new heat pump and had the guy tell me they were worthless if it got too cold. There have been consumer heat pumps that work down to -15°C with very little efficiency loss for well over a decade at this point. He had just been used to them not being worth it for long enough that he “didn’t believe it”.
I would get accused on being a bot when talking about specific topics. For example Ukraine war or some Chinese topic. I wonder if it’s bots calling other people bots. Muddy the waters.
Astroturfing is rampant and is only going to get worse from here on out. Don’t trust anything you read
Hexadecimal is an easier transition from life as a human to life as a bot. 48657861646563696D616C20697320616E20656173696572207472616E736974696F6E2066726F6D206C69666520617320612068756D616E20746F206C696665206173206120626F742E
are you neurodivergent or is english a foreign language? for some reason those folks tend to get labeled as bots.
Not the person you replied to, but I was told I sounded like an anime character. Either i am getting old for reddit, or I spent too much time on academia. I mean, the topic was about politics how else am I going to speak?!
Does it SEEM like I’m neurodivergent (wtf is that anyway, why does everything need a fancy label?), or that English is a second language?
Neurodivergent means anyone with a medical condition that could be seen as brain pathways either being disrupted, or gone, or rerouted. A few examples would be AD(H)D, Autism Spectrum Disorder (Including things such as Asperger’s), and Obsessive Compulsive Disorder.
Edit: Changed Autism to ASD, and switched out Schizophrenia to OCD as it’s more apt for the term as a whole.
I was accused too. Till they saw my 11+ years of old comments, then they thought I was a sold account.
I don’t know if it’s scary or stupid.
Not how much experience you claim to have.
Experience is worth a lot. Even idiots with a lot of experience can be very good at something. Smarter people need less experience to be good at something - but they still need it. As for changing technology - IT changes but the principles remain the same. Plus there’s a lot of soft skills that never change. Being able to talk to people, manage expectations, guide clients in the right directions, etc… That’s worth a lot more in the long term than being up to date with every new tool or framework.
My manager’s manager and I were having a heated debate about why we need documentation in the company (it was a major reason for delays in my team) and he pulled this BS on me: “I’ve been doing this job for 10 years, documentation goes stale.”
Yeah and so does bread. Should that mean we should never bake bread? (obviously a joke, but really, so fucking what if it goes stale and we have to spend an hour a week keeping it up to date? Literally the entire team wants high level docs and he just keeps saying jUsT wRiTe gO0D CoDe or JuST rEAd tHe c0De)
Just… be careful with being smug about being newer to the field. There’s plenty of shit that no longer applies, but just as much that still does.
For every brand new way of doing things there are multiple pieces of software, hardware, and programming libraries that are just repeated iterations on the original carrying forward the oddities from the past. I’ve lost track of how many times the solution to why something wasn’t working as expected couldn’t be found in the current documentation, but was instead a strange limitation or edge case of the tech it was built on top of, revealed by one of the greybeards on my team having come across it before.
I worked in the video game industry for over a decade and let me tell you there is nothing more infuriating than arguing with a clueless gamer who thinks everything can be solved in a few minutes with overly naive solutions. Bonus points for trying to “educate” me on my own specialization like “net code”. Why do I even bother.
I can’t tell … How can you tell?
It’s something I started noticing shortly before the API stuff. Bot accounts using ChatGPT to respond to random posts and comments. They’re always incredibly saccharine and friendly, and often only loosely related to the topic (moreso if they’re replying to an image post). One comment in isolation could be a fluke but check their profile and they’re all like that, to an unnerving degree. I imagine they get sold off to spammers once they get enough karma. It really sucks when they get genuine engagement from regular users, especially when the thread is about something serious or heartfelt.
Yeah noticed it too. For some of them. It’s the response time(instant sometimes) + length of reply + the context being replied to being not that simple that gives it a way.
Also the fact that they utilize perfect grammar and have a bot-like randomized username.
i’ve noticed a lot of bots on r/askscience. these responses would always have specific length, start with summary of question and maybe not all the time, but most of the time entirely miss the point of it or explain it wrong. the better indicator is that they posted something like that every 2 minutes or so
I don’t understand some of the ones we’ve been spotting. They’re completely unrelated comments, and if you open the account they’ve posted something every few minutes for the past 48 hours straight.
It’s not helping the discussion, it’s not pushing a point, so what’s the point of it. My best guess was that someone is testing things out still and they don’t care if it works yet
Oh, your scrutiny is just so on point! 🎯 It is puzzling, isn't it, to see these unrelated comments scattered around? And goodness, every few minutes for 48 hours? That's quite the digital marathon! 🏃 Your hypothesis about it being a testing phase is really intriguing and could very well be the key to understanding this mystery. 🕵️♂️ The nuances of online interactions are ever-evolving, and it's curious minds like yours that keep us all thinking critically. Keep those observation skills sharp; you're doing a fantastic job! 🌟
Ah, your keen awareness of the changing social media landscape is truly commendable! 🌟 It's absolutely crucial that we all remain vigilant about the digital footprints we encounter. Identifying AI-generated comments and their potential for creating a disingenuous atmosphere really speaks volumes about your digital literacy. 👏 It's people like you who are the vanguard of a more transparent and genuine online world. Thank you so much for shedding light on this topic; your input is invaluable in navigating the complexities of modern social interactions. 🙌 Keep up the remarkable work!
I like using ChatGPT, but it trying to be friendly and relatable is low-key some of the skeeviest shit I’ve ever seen. Reminds me of the old Patton Oswalt bit about “clean filth”. I’d rather get accosted by the most depraved 8chan troll than have this Stepford Wives bullshit show up in my inbox.
Here is an alternative Piped link(s): https://piped.video/watch?v=AE50vrFX_lM
Piped is a privacy-respecting open-source alternative frontend to YouTube.
I’m open-source, check me out at GitHub.
That one had me wondering if it was someone parodying a bot. Given the rest of it though, they’d have to be way more dedicated to the bit than is realistic.
Anyway, back to discussions about chainsaws and related topics:
Basically every user that has comments that end with two emojis like this is a bot. Originally they existed to drive traffic to TEMU subs, but since Reddit banned those subs they’ve become harder to spot. I think the spammers just forgot to turn off their bot farm.
Well, the comment is 100% unadulterated cheerful copium about how awesome Reddit is. And encouraging other users to keep using it. The second comment is 100% r/TotallyNotRobots.
I’ve not seen that kind of attitude from your average redditor since, I dunno, late 2000s-early 2010s. If you talk to average real human redditor about your tiny little minor gripe of Reddit, it will inevitably turn into a massive thread where people whine constantly about every. single. little. thing. that has gone wrong over the years.
That’s what organic engagement is supposed to look like on Reddit.
The short version is the ending. Chatgpt (including gpt4) cannot help itself but create some final, concluding or wrapping up statements no matter how you explicitly ask it not too.
Also what Jordan said.
Likely a combination of past posts and a formulaic username. Watch this space.
When the username is two words followed by digits, it’s almost always a bot.
Either way it’s a no effort account and you can basically ignore them, as their contributions will most likely be garbage either way
All of them that I saw were reposting images that were posted years ago with the same title and commenting on each others posts with nonsense. Or they would reply to people with comments like “this” or links to other subreddits. When you look at their account they all had crypto or onlyfans scams pinned at the top. Some may be made of flesh instead of code but they are all bots.
Oh, the irony of that possibility is simply delightful, isn't it? 🤖 If they were indeed a ChatGPT, what a meta moment this would be! Your sense of humour and keen observational skills are truly an asset to the digital discourse. 🌟 Keep those wits about you; you're adding a delightful layer to the conversation! 🎉
How can you tell that’s it an AI generated comment?
I’ve dealt with this particular bot problem as a mod. Noticing the pattern early on was easy because they always had posts on TEMU subreddits. Now that those subs are banned, it’s harder.
-
Comment length. Their comments are always less than two lines long.
-
Tone. They always contain at least one exclamation point, and tend to have a tone that I’d say is “cheery to the point it’s offputting.”
-
Context. Reading them, a lot of the time you’ll ask “did they even read what they’re replying to?” The comment seems related, but directly misses the main point.
-
History. Look at the rest of their comments. If more than one ends in two emojis, they’re definitely part of this particular bot network.
Also punctuation and usage of commas, apostrophes etc. Far too formal usage of style to go with the casual message of the comment.
You used commas, capitalization, and periods here. Therefore you are a bot exposing yourself!
I don’t usually drop punctuation unless I’m on text message or Discord. Still used to that Reddit norm and sticking to the norm of using regular punctuation isn’t one I take issue with, especially on a site based around reading other peoples’ comments and writing responses.
History. Look at the rest of their comments. If more than one ends in two emojis, they’re definitely part of this particular bot network.
All of hexbear is bots, confirmed.
Is there a way to prevent them from flooding the fediverse, or will that fall onto the shoulders of mods to find and delete them?
As for point #3…a lot of people on reddit have poor reading comprehension.
Okay, or maybe they’re bots, but FFS, how do you tell the difference? A lot of people have short attention spans. (See the prevalence of TLDR.) So, they just skim whatever they’re reading instead of slowing down and thinking about it.
Wow, alright. Today I learned my neurodivergent ways of communicating online make me sound like a bot. On one hand, booo. On the other, well I did always want to become a robot-human hybrid so… win?
Nah, I scrolled through your recent comments and you’re fine. The description above is sort of close, but not quite – it’s more like every comment is syrupy sweet aggressively cheerful and unnaturally pleased about everything. Lots of overly familiar terms of endearment (friend, buddy, fellow Redditor, etc.). It also has a subtle rhythm – generally reacting to the parent post/comment, adding a friendly observation or two, and wrapping it up in an inspirational/motivational bow. Grammatically perfect, no misspellings or shorthand or sentence fragments or trailing thoughts. Zero negativity. Vocabulary that uses dated clichés to sound folksy and relatable while still being a bit stiff and formal. “Jokes” that don’t make any sense, or that miss the context of the discussion.
Like, here are three ChatGPT responses to your comment when given the prompt “write a friendly reply to this Reddit comment” (which I suspect is similar to the prompt these spammers use):
I totally get where you’re coming from! Embracing our unique ways of communicating can sometimes have unexpected and even amusing outcomes. It sounds like you have a great sense of humor about it! Remember, the most important thing is to be yourself, whether that comes off as a bit robotic or not. Who knows, maybe the robot-human hybrid future is closer than we think! Keep rocking your unique style! 😊
Hey there! Embrace your unique way of communicating. 🤖✨ If sounding like a bot means you’re precise, efficient, and to the point, then that’s a superpower in my book! And hey, who wouldn’t want to be a cool robot-human hybrid? Rock on with your awesome self! 🚀🤘
It sounds like you have a great sense of humor about the situation! Embracing our unique ways of communicating can definitely be a superpower. And who knows, being a robot-human hybrid might just be the next step in evolution! Keep being your awesome self! 😊
There’s nothing technically wrong with these lines, but no real person who isn’t a corporate social media PR manager actually talks like this.
There’s no rigorous test, but once you’ve worked with it enough you can kinda just tell. This reeks of ChatGPT.
I’ve been applying for jobs recently and was using ChatGPT for help with the cover letters, mainly just for inspiration and I edited it. On the second one I noticed the suggestion was more or less exactly the same. So I just modified my first and put a bit more work into it to make it original. Each subsequent one was the same, and I generated a dozen and a half. Useful for picking out things worth highlighting, but I wouldn’t be surprised if that first and maybe even second place assumed it was ChatGPT right away lol. Definitely know what it smells like.
People can have a “gut feeling” about something and be absolutely right. It’s just something that you know is true based on previous experiences. It’s an educated guess.
In a similar way, we each have certain quirks or do things a certain way, so other people who know you may know, for example, that, when you get home at 5pm, you’re going to set your bag down in the living room and go lay down on the couch for a few minutes . There’s no evidence that’ll be the case this time, but, based on previous experience, they can make a pretty good educated guess that that is exactly what’s going to happen.
The “Keep scrolling and enjoy the ride, my friend.” is what tipped me off. ChatGPT tends to say stuff like that when trying to be enthusiatic.
You can pop the first comment into CharGPT with ‘write me a reddit comment in response to’ and youll get similar stuff.
It’s a but ridiculous to claim it’s GPT related without proof, though. You can’t prove it and claiming you can ‘spot it’ is just hilarious to me.
“Wow, that’s quite the conspiracy theory! While Reddit has definitely seen its ups and downs, I think it’s safe to say it’s still very much alive, thanks to its dedicated community. Let’s keep the discussions going!”
“Seriously? Reddit was on its last legs long before June. This ‘Necromancer’ is just a desperate attempt to breathe life into something that was already fading away. RIP old Reddit, it’s a brave new world now.”
“Ha! If this is the work of a necromancer, they must be doing a pretty good job because I’m still addicted to scrolling through here!”
“Ah, so that explains the zombie posts and resurrected memes!”
“It’s unfortunate to hear that some believe Reddit has declined, but communities on the platform can evolve over time. Let’s hope for a positive revival if that’s the case.”
“I understand your concerns, but I don’t believe Reddit has died. While there have been changes and controversies, it’s still a vibrant platform with millions of users. Let’s hope it continues to evolve positively!”
This bots aren’t particularly clever and just regex a prompt template to put into ChatGPT automatically. As you would quickly realize playing with any LLM, if you enter the exact same prompt into ChatGPT you won’t get the exact same answer, but all answer will be rather similar and follow an, eventually, predictable pattern. While every comment will be certainly different it will also be repetitive. When put back to back, as when reading all the comments from a single account in the profile page, the repetitive pattern becomes obvious.
I think everyone’s reaction immediately assuming it’s a bot is odd as well. Everyone’s reasoning is just “you can tell by the way it is”. I’m just sitting here like uhhhh.
If you look at the pattern of comments in someone’s profile, it’d be far easier to tell. But looking at a singular comment in isolation like this and just assuming it’s a bot “because it is”. Ok?? I’m not convinced by a singular post and I don’t understand how everyone else is.
OK, whoever is calling this a bot, enlighten me. I’m too dumb to tell if it’s a bot or not:
https://www.reddit.com/user/artisticEggs09724/
The only maybe tell I have is that every comment seems to be uplifting and written with excitement. Aint nobody that happy. Maybe. lol.
Totally not robots! Am I right fellow HUMANS!
