But Huffman said Reddit was now battling to ensure its users stay at the center of the social network. “Where the rest of the internet seems to be powered by or written by or summarized by AI, Reddit is distinctly human,” he said. “It’s the place you go when you want to hear from people, their lived experiences, their perspectives, their recommendations. Reddit is communities and human curation and conversation and authenticity.”
As Reddit becomes an increasingly important source for LLMs, advertisers are responding with what one agency chief described as a “massive migration” to the platform.
Multiple advertising and agency executives speaking during this month’s Cannes advertising festival told the FT that brands were increasingly exploring hosting a business account and posting content on Reddit to boost the likelihood of their ads appearing in the responses of generative AI chatbots.
Starting when?
That ship has sailed.
I know he’s just selling the platform to AI companies, but it’s an odd take considering they’ve been moving away from being a message board and towards being just another content feed for years now.
Looked up some information a bit ago and found Reddit using AI written answer pages to aggregate information. Which- why? The information was already indexed by a search engine, the sole reason I landed on this page. Why are you offering me secondhand, watered down information written by your idiot machine when I can instead read sourced accounts with citations?
It’s interesting to know that you can do this, but I could not care less about a machine’s perspective on what is or is not the prevailing opinion. Moreover, it completely contradicts your stated goal, you perjurious pool of feculence.
deleted by creator
The ship was never at that port.
Reddit was using bots and vote manipulators back to the Digg days, as well as scraping/reposting akin to what 9gag became famous for.
Anyone who participated in any r(slash)place could tell you this has been a lie for at least a few years.
Really? So he’s going to get rid of all the bots?
Yeah, they may not have AI generated content (yet) but I left because their AI moderators REPEATEDLY warned/banned me from the site. Every time I sent a protest letter and had an actual person look at the post, I was unbanned, but I am fed the fuck up. That site is a study in enshittification. I am done with those assholes.
And, if they think I’m going to let them scan my retina in order to post, they can fuck right off.
Not by this human.
If we don’t look for bots, there aren’t any. Sounds more republican every day. If we don’t record covid deaths, there aren’t any.
Any text based sub hitting all regularly is full of AI slop and sensational stories. Wouldn’t be like this is Aaron were still around.
Fuck, can this loser just fuck off into oblivion already?? You got your money pig boi now piss off!
It’s like Zuck. Why do you keep at it? You won already. The answer is that they are deeply flawed people who needed to go into therapy decades back.
Yeah there’s something deeply wrong with them
He didn’t say entirely or how many, lol. If there’s 2 humans writing, it’s technically still, “It’s the place you go when you want to hear from people.”
Exactly what I thought, cause a lot of comments are from bots and even the ones that appear from humans aren’t necessarily humans. How can you say it remains like it is now and it is written by humans at the same time?
Does it count as “a human” if it’s a bot reposting human-made content for the 50th time?
I’d say the content will forever remain “human written content”, but also that it being once written by a human doesn’t mean the repost is a valuable addition to reddit.
The site is loaded with engagement bots, and that has increased exponentially in the last year. Before I left, I was constantly getting pinged with inane questions to try to elicit a response (engagement). So much different here.
…with AI admin and moderation. Anyone who’s been banned from reddit will tell you that, hell it says as much in the message they give you. I was banned once for saying “slava ukraini” which according to my ban was a direct attack at a minority (apparently russians are a minority according to reddit bots) and then another for asking how long before citizens begin shooting at ICE agents which, again according to reddit bots, was inciting violence on a minority (ICE agents are now also a minority according to reddit).
the vast majority of posts and comments are bots. the obvious reposts with misleading titles to generate karma to allow said bots to have enough karma to then post on various other subreddits to push whatever bullshit agenda or product they want to push.
The twist - AI was originally trained on human input, so it’s human even with 99% bots, from a certain point of view.
You will find that many of the truths we cling to depend greatly on our own point of view.
LOL!
meh, haven’t been to reddit in months. nothing but bots arguing with bots. won’t miss it. /fuckspez
And controlled by 15 mods.
Reddit is filled to the brim with bots.
He can’t let people poison the training data, or else his site will have no value to its customers.
how cn we poison the training data for llms? only ever saw stuff for image gen-
imfo: i updated the spelling. it was unreadably bad before.
how cn we poidon the trsining data for llms? onli evr saw stuff for image gen-
That’s how.
oh god >~< i actually didnt realize autocomplete wasnt working at all heheheheeee…
i… will fix up the message, this is actually unreadable… did not mean to go that far >v<
If LLMs train on text output from LLMs, the results will degenerate into total garbage over time. The people that buy reddit data for LLM training know this. They will stop buying if they think there’s a lot of LLM text on Reddit.
True, but this is an advertisement for something that doesn’t exist.
It’s gotten to the point that if someone were to ask me what my super power would be if I could choose, I’d say that I’d like to be able to see what is human and what isn’t when I’m online. Very boring super power, but it would make some things a lot easier.
Ironically I also lost count on how many times I was accused of being a bot near the end of my time on reddit. I wouldn’t be surprised if those who accused me were bots themselves. It was just insane sometimes.
But good luck with that, Mr CEO. I’m sure you care very much about your human users on your goofy platform.
Me too…any tips for spotting it?
It is so tricky and i dont have any good answers as I’m a tech-idiot, but I’ll try my best to give you some suggestions:
In the past you may come across a comment that sounded overly robotic and emotionless when they commented, but that might as well just be someone on the spectrum. Nowadays some bot comments sound more human than actual human comments. I really cannot tell.
I think I was often accused of being a bot because I tend to go into detail about things that interest me and try to explain things in too many paragraphs, lol. I have tried to learn how to shorten my replies to people but I fucking fail everytime. I literally just spammed a friend with a long ramble about why the Syrian flag looks different and went into the history of the flag and made a cliff notes version of the conflict in Syria because it randomly became an interest for me for a few minutes after a car passed my window with the new Syrian flag on the car helmet. So maybe that trait is a bit off putting to random people who cross my path online. I dunno.
Anyway, I have heard that some people look at sentence structure and grammar to decipher who is real and who isn’t. If the grammar is too good, the sentence structures too perfect, then some people suspect it is a bot. But it’s probably just a matter of time before bots will adapt to poor grammar and writing patterns. If they can go from sounding like a Wikipedia page to cracking jokes and sounding human in less than five years, they can learn how to imitate shitty writing patterns that most of us have. I for one tend to accidentally press the s when I want to press the a and I usually catch it before posting, but sometimes I overlook it.
Usually, on reddit, I would check the comment history of someone I suspected of being a bot. If they posted way too much, like seconds between each comment and in various different subs for hours, I would conclude they were a bot, but I don’t think all bots are like that.
I dunno how I would spot bots in the fediverse, tbh. Maybe similar, checking the comment history if i suspect a bot, but otherwise i dunno man.
Usually I go by how people behave. Anyone who acts insane or aggressive gets blocked. I am too old to deal with shit like that, but that has less to do with bots and more to do with no longer tolerating extremely divisive behavior.
That’s really helpful thanks mate. And I agree, banging your head against a brick wall just hurts your head
