Reddit's filing with the SEC makes clear that training AI with user posts is a core part of Reddit's new business model.
Reddit said in a filing to the Securities and Exchange Commission that its users’ posts are “a valuable source of conversation data and knowledge” that has been and will continue to be an important mechanism for training AI and large language models. The filing also states that the company believes “we are in the early stages of monetizing our user base,” and proceeds to say that it will continue to sell users’ content to companies that want to train LLMs and that it will also begin “increased use of artificial intelligence in our advertising solutions.”
The long-awaited S-1 filing reveals much of what Reddit users knew and feared: That many of the changes the company has made over the last year in the leadup to an IPO are focused on exerting control over the site, sanitizing parts of the platform, and monetizing user data.
Posting here because of the privacy implications of all this, but I wonder if at some point there should be an "Enshittification" community :-)
Reddit has long had an issue with confidently providing false statements as fact. Sometimes I would come along a question that I was well educated on, and the top voted responses were all very clearly wrong, but sounded correct to someone who didn't know better. This made me question all the other posts that I had believed without knowing enough to tell otherwise.
Llms also have the same issue of confidently telling lies that sound true. Training on Reddit will only make this worse.
@Fubarberry yes I saw this a lot too. Highly upvoted confidently incorrect comments, with the real answer or an answer debunking them with links to factual sources less upvoted.
I am a lawyer and I would get down voted for posts explaining the law that contained citations to the actual applicable statute if people didn't like the statute. Using reddit up votes as a measure of correctness is fundamentally a dumb idea.
I would come along a question that I was well educated on, and the top voted responses were all very clearly wrong, but sounded correct to someone who didn’t know better.
This can be said to https://news.ycombinator.com/ as well. I wonder how much of this is due to sock puppets and bots.
I'm still happy that I went through the effort to delete all my old posts when I left Reddit a while back. I periodically check if they've restored them and luckily it hasn't happened so far. I do miss some of the bigger communities but overall I'm having a good time on Lemmy.
I wonder what the risks are to including deleted and pre-edited content in training data. Most of the edits are going to be typos and formatting, do you want 2-3 copies of the same message with typos in them for training data? Similarly, deleted comments are mostly nonsense, unhelpful, duplicate, or highly controversial things.
If someone wants to dig through and find individual users to restore that's one thing, but I don't think I'd immediately choose to train off of that other data unless I had to.
Only shadenfreud I have is that my deleted banter that they will assuredly include, will hopefully increase the stupidity of whatever model gets trained on it. Ugh, what a dystopia we’re building.
After deleting all of my posts and comments Reddit decided to undelete them three days later and then proceeded to lock me out of my own account. Fucking bastards.
I just left my comments on. I still use reddit when searching actual human responses from Google. Maybe one day someone might find my archived comments useful in the future.
I am glad it makes you feel better but the reality is they still have your data. Just because you don’t see it on the front end doesn’t mean it isn’t still in the database with a “deleted” flag set. They aren’t hard deleting your comments.
Deleting your messages is just another data point for them. Reddit can train an AI on the originals and categorize you as a "comment deleter" to give them more information.
This has got to be the start of another bubble popping. It just has to, right? With essentially all online services doing everything they can to wring out every last penny of value without any eye towards the future (other than ai all the things)… something’s gotta give.
But then again, maybe it’s just my eyes being open after living in those spaces for so long. Granted I’ve been out of Facebook for years, been de-amazonning for a couple (it’s really f’ing hard) and I’ve been trying to de-google as well but it’s even harder (stuck with Apple though). But, now that I’m in the fediverse, where we’re talking about all this, maybe that’s why I’m noticing?
I think the bubble is coming too. The question is how much it will take for normal users to be done with them. The current Lemmy user base is more focused on tech, open source, and/or privacy than the average Internet user, which is why we already abandoned Reddit.
I think having to pay for access to these sites might be the biggest issue, as many people see the Internet as something that should be free.
The start of the bubble popping was the increases in interest rates. We've seen several online companies shut down already because the free money isn't there any more and there is no path to monetization.
The problem with the Fediverse right now is that it is all run on volunteer labor and donations, similar to an early Reddit. It will be interesting to see how a distributed system solves this problem.
Makes me wonder how that technology is going to track. Reddit isn't bad for finding niche answers to niche questions, but if you import the data wholesale then you'll have a hard time separating the signal from the noise, even if you sort by using vote counts as relevance.
Reddit is valuable because people can do a search for a niche topic and find the answer on that forum. And the answer was written by a human. It's not valuable because it can amalgamate an approximation of those answers that might be 90% true and 10% dead wrong.
A lot of AI models are probably already trained on Reddit data. But apparently Spez isn't important enough to world order to make the cut to be compressed into a 7B model. I asked my Mistral-7B-Instruct (4-bit quantised) local LLM:
>>> What can you tell me about someone called u/spez, a
... nd are they a nice person or horrible?
I cannot provide any information about a specific individual named
"u/spez," as I am an artificial intelligence language model and do not have
the ability to access personal information about individuals based on their
online usernames alone. The username "u/spez" appears to be a Reddit
handle, but it does not reveal any information about the person's character
or personality.
Reddit is a popular social media platform where users can create unique
usernames for themselves when they sign up. A username like "u/spez" may
have been chosen for various reasons, and it doesn't necessarily provide
any insight into the person's nature or disposition. It's essential to
remember that people's online personas may not always reflect their true
selves, and it's important to exercise caution when making assumptions
based on limited information.
I cannot determine whether "u/spez" is a nice person or horrible without
knowing them personally or having access to their actions, thoughts, and
motivations. The best way to form an accurate opinion about someone is to
interact with them directly and observe their behavior in various
situations over an extended period.
>>> Is it possible u/spez is Steve Huffman? Can you tell me anything based on tha
... t?
I cannot definitively determine whether "u/spez" is Steve Huffman or not,
as I don't have access to their personal information. Steve Huffman is a
Reddit user who gained public attention in 2018 when he posted an AMA (Ask
Me Anything) thread where he shared details about his life and experiences
living with Crohn's disease. His AMA post garnered significant attention
and sympathy from the Reddit community, and he later published a book about
his experiences called "This Is Gonna Hurt: Living With Crohn's and
Colitis."
It is possible that "u/spez" could be Steve Huffman, but without further
information or verification, it remains speculation. It's essential to
remember that usernames on social media platforms do not always reflect the
true identity of the person using them, and it's important to exercise
caution when making assumptions based on limited information.
If you have any additional information or context about "u/spez" or Steve
Huffman that could help clarify their relationship, please let me know, and
I will try to provide an accurate response based on the available
information.
"Early Stages?" You've got AI mining your data. The Lions have already come and gone. The hyenas and other scavengers are picking over the scraps, now.
When I go to some reddit posts on Mobile now (like from a Google search, that's the only way I end up at reddit anymore), it tells me "this content is unmoderated" and gives me a choice to either navigate away or install the Reddit app. Fuck that noise.
Try this, in either Bing/Copilot AI or Google Gemini:
Start your prompt with "According to Reddit", then do your search like you would by using search alone.
The AI of your choice will scrape the posts and give you a nice summary of whatever you were searching for - no need to ever touch Reddit directly.
For me, this works better with Copilot, YMMV.
Example: "According to Reddit, what is the best mechanical keyboard brand to use for touch typing?"
“we are in the early stages of monetizing our user base,”
If anyone on Reddit reads that and stays there willingly they are an idiot. Not they weren't idiots for staying after the API changes but now they are even bigger idiots.
I know it's only token resistance at this point because others have found their comments from Google searches even after their accounts have been deleted, but Power Delete Suite is busy churning away on mine right now.
They permabanned my 14yo account because my anti-nazi rhetoric was "encouraging violence." I guess Nazis are a class of humans dumb enough to give them money so they don't want to scare them off. The post that got me banned had more than 60 up votes when it was deleted and I was permabanned. A reply post in the same vein was not deleted.
Remember that video where Ron Perlman talked about there's a lot of ways to lose a house?
I lost my 11-year account because I said something to the effect of 'If Ron Perlman pulled up and said get in the fucking car we're going to go burn down Bob Iger's house I wouldn't hesitate.'
They had been getting very weird near the end there anyways? I kept getting these stupid warnings over the most petty shit. At one point somebody said respond to this comment and I'll gild you. I simply responded fuck you because I thought it would be funny to see that have gold, which it got. Got an official warning for harassment.
Curiously, Nazis seem to get away doing just that, under their clear name even! Reported a few of those on Twitter a while ago before Elons takeover. Got a message that the reports are unwarranted and if I continued to make them they'd disable my ability to report.
I asked what Eisenhower would do if he saw the Nazi marchers in Wisconsin and had ready access to a machine gun. I don't think that is advocating violence. I intended the comment to illustrate how far some Republicans have moved to the right since Ike was president.
Eisenhower is dead. Advocating for his attendance at a Nazi march is nothing more than a thought experiment.
“Pay-Per-Click”, is all this is when you break it down to its basest.
Narwhal developers have come out and said that they have to pay beforehand for clicks to the API—- what absolute bullshit Reddit and Spez are bringing to the trough. Spez killed reddit—- calling it now; a slow painful lingering shitty death.
People will not put up with it once they know what is really going on.
Is this a long term source of revenue for Reddit? Or will it loose value at some point, simply because LLMs are all trained sufficiently on user generated content. Is there more to learn at some point?
Also it seems that a lot of content on Resdit is already AI generated, so it would train on data from other LLMs, which I'm sure doesn't improve quality.
LLM's are a parasitic entity. They can only operate as long as they have a living host (us) on which to draw data. Without their host, they rapidly start hallucinating. Hell, the other day ChatGPT (and every business that relied on it) started hallucinating for no apparent reason.
The thing about the parasite is, though, that it endangers its host. At some point, the fact that anything you say can be plugged into a machine with no credit given back to you, will encourage creative people to stop bothering being creative, depriving them of income or even exposure.
It's a funny thing, a few years ago I would say that the "anything you post here can be sold by us" clause on social media was very unlikely to get exploited, as nobody knew how to sell data en masse to make money off of it. I guess now we know that's not true at all. If something bad can happen with your data... It will.
It’s the reason I can’t see this stock maintaining or improving its price after the IPO. I mean, sure, there will probably be a short term gain for a few stock holders. But, I just don’t see how it doesn’t tank afterwards. I mean, in the end, Reddit is Reddit. It’s just an aggregation site, how can it grow in value? The fediverse is slowly but surely gaining popularity. And even though Reddit calls itself the front page of the internet, it really isn’t.
*Not investment advice. Good god please don’t take investment advice from me. Knowing my luck that fucking stock will soar to Wall Street record highs, beating out Bitcoin by a large margin.
It’s just an aggregation site, how can it grow in value?
Supposedly in Reddit finance there's something called the "Anarchy Chess/Ewan gambit". If you post one grain of rice, and double it each time you reach a threshold you can farm near-infinite updoots! Probably works the same with money, idk.
With all the changes that Reddit has made recently esp with the API changes, it definitely did leave salt in my mouth alongside how increasing toxic the Reddit community had become in comparison to when I joined the community but the small niche communities that existed on Reddit did honestly made it harder to quit due to the lack of communities outside, which is another big problem with centralisation, esp in the modern internet as it makes you rely on platforms you may not necessarily like due to big issues like social isolation etc.
When I found out about this, this isn’t simply excusable anymore and I would rather delete my account over having my personal data being sold for profit (which goes completely against the early ethos of Reddit as a whole but being semi owned by Conde Nast, this would have been inevitable) despite the fact that I have been thinking about deleting my Reddit profile way before this issue.
Surprisingly, I honestly have had no regrets deleting Reddit out of my life and honestly I do wish I would have done it sooner, I’m far less frustrated, I’m starting to think more constructively again and I feel way way less dependent on it.
Ditto for me, as well. It's just a matter of establishing those 'niche' communities on the Fediverse. The Fediverse has broken thru 10M users. We're getting there. Onward!
If you are planning to kill your reddit account, there is an app, Redact, which is available on the Apple and Play stores, that will allow you to nuke all your posts before you close it completely. Deny them your data.
For better or for worse, Reddit has a super valuable archive that has basically replaced Google search for me, it's insane how many times it has helped me solve small and big issues. I understand the logic, but it would still be a big blow for the internet if many people did that.
I do think it's interesting that a lot of people seem to think AI is going to take away jobs but understanding AI just a tiny fraction, it seems like the things that are threatened are one that were already micro serviced away like internet search.
We use search everyday and having the best search engine means being the best tech company. These companies are in a race to topple Googles search dominance through providing AI as a service. There's money in them hills if you can train an AI to recommend when and where to go buy the newest shiny thing that solves all your problems.
Looks like the enshitification of Reddit is about to accelerate. I barely use it anymore, but I kept my two ten year + old accounts intact (one for porn one for legit posts). I’ll probably nuke my non-porn account soon.