This is just the practice of invasive data harvesting taken to its natural conclusion, which any tech savvy user or computer geek brought up in the pre-postmodern social media era of the Internet (before Facebook) could have seen coming decades in advance. The only winning move here is not to play, accepting the unfortunate consequence that you will need to "limit the reach of your profile just to avoid participating in the new technology" - because nothing on the Internet is ever forgotten, and can (and will) be used for any purpose.
Adding a few magic words to bewitch the AI into not scraping your profile is the new superstition of the digital era, a cousin to the pseudolegal "no copyright intended" incantation often seen on pirated YouTube videos of yesteryear. You cannot have your cake and eat it too, for there is a fundamental tradeoff between privacy and convenience as popularized by Schneier, 11 years ago. [0] You must stop using the platform and do something other than continue to consume vapid social media nonsense; yet no one ever listens or cares, for the revealed preference of the masses is to continue to not be users of the system but to be used in exchange for "free" access to these platforms.
[0] https://www.schneier.com/blog/archives/2013/06/trading_priva...
I'm not sure why having an LLM use my output in particular is problematic - I'm pretty sure OpenAI trained GPT-3+ on every single bit of data they could find, so they probably included a bunch of stuff I've written here already (though it is minuscule fraction of their vast corpus of course). Losing anonymity is something I'd be much more worried about overall (not that it's entirely separate I don't think it's the same).
There’s many good reasons, but for me it’s that I don’t want companies profiting off of my posts that have no intention of profit sharing.
When I make a Youtube video and companies run ads on it, I get a piece of that pie (assuming I meet the requirements, etc.).
That same video fed into Gemini so google can charge for AI video generation? I get nothing, Google makes bank. As a user I can pay for YouTube premium and not see ads, but as a creator there’s no amount I can pay to not feed Gemini.
>That same video fed into Gemini so google can charge for AI video generation? I get nothing, Google makes bank.
How do you feel about commenting on this site for free, which probably provides some benefit to ycombinator?
I would happily pay a reasonable monthly subscription to this site or similar. No problem paying for a service that treats users with respect. Also using this site has taught me all sorts of things that likely made me some money indirectly. It seems mutually beneficial without selling my data or paying for HN.
That said, if YC made a deal with some tech company to give them the firehose of data to train AI, I’d probably stop using HN. I stopped using reddit for a similar reason despite being a very frequent redditor with like 60k karma. I know it’s all pretty open and getting fed into many different LLMs anyways, but thats not necessarily YC’s fault.
My ideal would be strong government regulations regarding AI training, requiring explicit opt-in that isn’t buried in a ToS or EULA. Ideally companies would require a “non-AI feeding” version of their website to legally run in my country.
I can’t imagine a scenario where this happens in the current system, but I sure can fantasize.
Right, but are you objecting to AI training because companies are benefiting while you're being uncompensated, or think AI training is fundamentally bad? Your previous comment suggests it's the former, but by the same logic you shouldn't comment online either, because that also benefits the company and is uncompensated.
I get the satisfaction of telling off strangers
> for me it’s that I don’t want companies profiting off of my posts that have no intention of profit sharing
I won't argue with your position, but most people would be hurting themselves more than they would hurt the companies, even in absolute terms.
> I don’t want companies profiting off of my posts
I despise this attitude. It's so entitled.
Our history of forever extending copyrights and protecting "intellectual property" has run amok, to the point where the average person thinks their scribbles, utterings, and ideas are valuable enough on their own to be worthy of a pay day. It's the culture of "My cut, my cut, my cut!"
Someone else profiting is not a tragedy to get up in arms about. The fact that you were somehow, tangentially, kinda sorta in the vicinity of that profit does not and should not mean you are owed money.
If you want to talk about privacy, sure, that's an issue worth bringing up. But, "I'm only mad because someone else made money and I didn't get paid," has nothing to do with privacy. It's pure greed, entitlement, and envy.
You know what? If you want to profit, do something to create value. Write a book. Start a paid newsletter. Create a startup. Put on a show and charge admission. Nobody is stopping you.
But if someone else figured out how to use a snippet of a comment you made 10 years ago as one-quadrillionth of the training data for a powerful LLM… if someone else figured out how to use your publicly-shared social media posts to attract advertisers to a platform they built… if someone else used 6 notes from a song you once sang to create a smash hit… kudos to them. They created something of value. You should've and could've done it yourself. Hell, you still can.
But nobody should owe you money. We should not have a society where people who actually create stuff are subject to endless friction and threats from do-nothings and patent trolls demanding "my cut" if the metadata from their words or actions contributes to 0.0001% of someone else's idea that they turned into profit with hard work.
> You know what? If you want to profit, do something to create value. Write a book. Start a paid newsletter. Create a startup. Put on a show and charge admission. Nobody is stopping you.
I believe that the GP's complaint is that their content online is actually being scraped and turned into value for companies, they would want compensation for it.
I'm personally of two minds on this, posting public content online includes no guard rails for how its used. I also disagree strongly with LLM companies throwing mountains of resources at scraping the web though, if nothing else it feels very much like a monopolistic play leveraging massive power in those resources to create a competitive edge that other players couldn't compete with.
I was served a youtube ad for a service that will let you pick a topic to write an ebook on, write the book for you, give it a layout design and a cover, then you can market it. A bunch of topics available. I find this to be gross. Look for an explosion of crap on Amazon bookshelves.
Do you want me to publish 12-15 different ebooks containing the content you actually worked to create, and it just found permutations for?
> Do you want me to publish 12-15 different ebooks
Just make sure you spread it out over a few days and you're all good: https://arstechnica.com/information-technology/2023/09/ai-ge...
It's not possible to not play. Even if I chose to avoid social media, my kids' school insists everyone interest with their teachers through some crappy branded app.
Read the TOS. Didn't like what I saw. Told them I didn't want to install an app on my phone. Asked nicely if there was some way I could participate without it. They acted like I beat my kids.
Organisations insist that I interact with them using an app. I show them my Nokia feature phone and they find an alternative.
Really, this is more damning. They could accommodate you but they don't because you could submit to their bullshit and choose not to.
Maybe the answer is to carry a dead feature phone to show to them. Like a talisman to ward off their evil.
How do schools actually manage requiring this? Is it required that all students have access to smartphones, tablets, the internet in general, etc? If a student doesn't have a device to install the crappy branded app on, what does the school do?
I don't have kids yet though that time is likely coming, this goes firmly on the growing list of reasons why we'd choose to homeschool.
Schools that do this typically provide a Chromebook or iPad to each student.
When you homeschool you will likely buy said crappy app (or another one) to help you actually school your kids...
Sure it is, you can homeschool for example.
No thanks, don't want to put my kids through my childhood
Assuming you were homeschooled and it went poorly for you, was your takeaway that it could never be done well? What did you find that makes homeschooling always worse than trusting the local, likely poorly paid and appreciated, state educators?
> Asked nicely if there was some way I could participate without it.
You've done your duty with them. Now write a letter to the school board. Give them a little while to respond, and if they don't, start handing copies of the letter to other parents outside of the school gates.
If you can't get other parents to care, then you're in trouble. Try to claim a religious exception.
The outcome is that spaces which are prone to AI slop and spam will wither and lose real users, eventually collapsing on themselves. For the social media platforms, good riddance
What I'd give to see the user retention stats at DeviantArt ever since they allowed it to be filled absolutely wall-to-wall with the most boring AI art you've ever seen. Not just allowed but encouraged by jumping on the bandwagon and integrating their own generator. Anecdotally it feels like there's not just an insane glut of AI slop drowning out everything else, but also less "everything else" over time due to user attrition.
It really can't be overstated just how much it's come to dominate the entire site, skimming the frontpage now I immediately spotted a generic Midjourney-core image and the creator has posted 4100 pieces to date after joining... two months ago. 68 uploads every day on average.
100%. It's intensely frustrating that they didn't at least firewall that vast amount of AI noise using a subdomain or something (e.g. ai.deviantart.com).
They do have an AI tag that creators are supposed to set where applicable, and there's an option to hide everything with that tag, but the enforcement is weak so there's a lot of obvious AI spam which isn't tagged. Many of the AI spammers are using DAs monetization features or promoting a Patreon so they're incentivized to not set the AI tag to maximise their reach.
Deviantart was already noise of unappreciated forms of expressions
People just use it as a hosting platform
Sounds more like a UX issue that you can see galleries
https://en.wikipedia.org/wiki/Gray_goo
This made real.
When I pop open facebook and see some page has somehow gotten onto my feed with an AI generated children's science fair project I always notice that it has 60k+ likes. To me it says that the stuff is indistinguishable from real for enough people that discerning users are just helplessly along for the ride on every platform.
The article purports that the reason slop is here to stay is because people like it. The next iteration, IMO, will be AI generated children's science fair projects that also have a pepsi logo in them, still with 60k likes.
We need to move above that before it kills the web though - if the end state is that you want to look at any old random adorable puppy pictures so you just ask your machine to generate them for you then we might have a way out of this internet death cycle. I'd hope that if you can trivially produce your own better tailored slop on your local machine then the AI slop online will lose any value - I'm concerned that we're going to lose a lot of good content created by artists before we reach that level though.
I suspect the actual outcome will be a rise of manual curation and providence where a feed of adorable puppies or discussions on technology will rise or fall by how diligent the moderator is at keeping slop out of the feed.
> When I pop open facebook and see some page has somehow gotten onto my feed...
There are still people watching Facebook with unfiltered feeds? There are filters for that.
From the article: "The main people benefitting from the launch of A.I. tools so far are not everyday Internet users trying to communicate with one another but those who are producing the cheap, attention-grabbing A.I.-generated content that is monetizable on social platforms."
Yes. The main use case for LLMs remains blithering.
> The main use case for LLMs remains blithering.
And in a close second-place, to cheaply/easily counterfeit signals that we are/were treating as indicators of human things like "intelligence", "time-investment", "emotional involvement", "distinct identity", etc.
I wonder how many traditional cover-letters / essays will vanish because the format has been so debased it doesn't mean anything, and people will say: "Just give me the core bullet-points."
> notice that it has 60k+ likes. To me it says that the stuff is indistinguishable from real for enough people
Or that buying likes from bot farms is incredibly cheap.
And how many of those likes are AI-generated?
Or humanity will devolve into confusion over what is real or AI generated for the next decade until we learn to live alongside the monster we have created.
Discord and heavily moderated forums are the future.
IRC still is around. Pretty sure most of the user base is human still.
Lots of bots. Also body-only-soul-less users (as I call those that never look at messages).
Yeah, small, invite-only "walled garden" style communities are definitely the future of the web.
There has always been those of course, but I think they will just end up becoming the default, rather then the fringes. And the thing with these smaller communities is that they are far less tolerant to their hosting platforms pulling dumb shit (like stating they will train AI's on the comunities content).
We can only hope
I wish this was actually true. The unfortunate reality is that the average internet user can't tell that they're looking at AI generated content with an AI generated caption and AI generated people posting AI generated comments, as we've already seen on Facebook.
Hard to make any forecast about "spaces" "prone to AI slop," whatever that means, besides social networks that you personally do not like, in aggregate or individually.
TikTok and Meta apps are definitely growing faster, in relative or absolute terms, than the New Yorker and the New York Times audiences are, paid or unpaid. If you compare aggregate social media to aggregate news and magazines, the latter is shrinking, if you exclude their presence on social media itself.
If anything we already live in a post-scarcity world with regards to engaging content, starting a few years before the advent of generative AI. Why would AI generated content reverse that trend? Isn't TikTok's feed algorithm agnostic to whether a video is AI slop or user created? Isn't Instagram's? Are you really going to play No True Scotsman with "spaces" "prone to AI slop?" Shouldn't Kyle Chayka, who supposedly wrote a book on this, know that? Aren't there already too many good books, movies, TV shows, video games, operas, plays, etc. to consume?
The toughest thing about this article is it is longing for a world that hasn't existed for a long time. If anything, the New Yorker and the New York Times, by doing a bad job at being media companies, have reduced the amount of new narrative creative projects that can thrive, not increased it. They never look in the mirror. The fickle and sometimes vindictive personalities that work there are not allied with narrative creators.
The idea that discovering one or two diamonds in the rough offsets the incumbent cultural trends the New Yorker reinforces has long been dead - there is just way too much new stuff for any traditional media company to accurately review, report on and amplify. Everyone thriving on YouTube, Instagram, Steam, TikTok, hawking their shit, figured that out.
It's even crazier to me that Kyle Chayka, who wrote a book on this, misses the mark here - I mean he should know about non-negative matrix factorization, which basically was the beginning of the end of traditional media, he should be able to make the leap that invention that enabled accurate collaborative filtering at scale killed The New Yorker, not AI, or anything in between 2000 and now. He should know there's absolutely no reason that AI generated content would be treated any differently than any other bad content, by NNMF or whatever feed algorithm.
There's a possibility that the reason the NYTimes and Conde Nast have to reinvent themselves is because they do a bad job. To them, "A.I." is just another effigy, when the reality is that fewer and fewer people care what important New Yorkers think. Listen guys, it's not looking good for the writers, better to get your head out of the sand.
> TikTok and Meta apps are definitely growing faster, in relative or absolute terms, than the New Yorker and the New York Times audiences are, paid or unpaid.
Imagine the doctor walking into the room and saying "Congratulations! The tumor is growing faster than your legacy cells."
Yeah, I don't think this is a good thing. It's just to say that Kyle Chayka and this HN commenter don't offer any remedies. They go and complain and mock people's superstitious Instagram shitposts.
If you care about narrative creative media, the best thing you can do is pay for it. Whatever that means to you. That is my remedy. The New Yorker isn't going to go out and promote Substack, and the deeper you think about why, the more you realize it is the New Yorker who are the assholes.
I don’t have a remedy to offer for the problem at large, which is why I didnt suggest one, but on a personal level I would read the New Yorker, which is high quality content, imo
> If you compare aggregate social media to aggregate news and magazines, the latter is shrinking
This isn't even factually correct, the New York Times has done well, and grown over the last decade. In particular their move in the early 2010s towards paywalled premium subscription content, away from advertising, which was lambasted at the time, was in hindsight a very smart move.
Yes, in the aggregate premium offerings don't come close to the size of the market of slop, it's always been a numbers game, but if you're talking about cultural trends, the people who run those slop factories read the Times and the Journal, they don't watch Youtube shorts themselves and probably keep their kids a mile away from it.
Cultural capital and literal capital aren't the same thing. Danielle Steel has made 800 million selling 200 smut novels, but that hasn't given her a lot of cultural status or influence. The people who make decisions and set trends don't read her books.
If that was true Twitter would be collapsing. The vast majority have been too passive to leave. I no longer think there's a threshold of worsening that will disrupt Twitter, Reddit, etc.
> For the time being, though, avoiding A.I. is up to you. If it were as easy as posting a message of objection on Instagram, many of us would already be seeing a lot less of it.
It's true that it is quite hard, but there are ways to reduce it for sure. Here is what I have done:
1. I've deleted accounts for websites that promote AI. I have already deleted LinkedIn, Github, Medium, and a few others.
2. I have stopped supporting businesses that use AI/support ones that are against AI. For example, the company behind the Procreate iPad app is 100% against AI so I support them. Also, in my professional life, I have already refused to collaborate with three separate companies due to their promotion and use of AI.
3. I've deactivated any tools that could be AI based like assisted writing tools in Gmail.
4. I do not click or read any articles with AI-generated images or text. Nor do I watch any videos any more.
5. I am reconnecting with friends and share with them through email and other means.
In my opinion, the internet has gotten WAY worse with the introduction of generative AI. Generative AI itself is not the root cause of course: the aggressive capitalistic takeover of the internet is, but AI is the apex tool for that and it makes the internet a rather horrible place.
A commendable effort (those are some serious sacrifices/actions - no videos!?) but it's a bit at odds with posting about it here which puts into question how much reduction in AI content specifically is occurring vs how much reduction in general consumption (which isn't necessarily a bad thing though). I.e. Y Combinator invests in hundreds of AI companies, >25% of the front page posts are regularly about AI, it was an original investor in OpenAI, users here regular post and discuss about how the latest AI models are even harder to catch. Even if you try to use the site in the most human way possible (filtering explicitly AI content, not engaging with comments referencing it, etc) it still seems like there would be a ton of AI content at a place that does more with AI than most of the companies listed.
All that isn't to scare you from Y Combinator or anything, it's just to get your thoughts on how much you think these actions have really changed your consumption ratio vs your overall consumption? And how have you come to any certainty in the measurement of e.g. whether an article you're about to click on has AI images or text (that would seem near impossible to do much of the time even after reading?)
> I have already deleted [...] Github
Given that Gitlab and GitBucket are also integrating AI, is getting out of these two platforms even a realistic choice for 99% of devs working for a regular company ?
On the other hand, convincing your CTO to get your organization out of any major code management platform feels like a pretty exciting chalenge to tackle.
Eh. Gitea is a thing. I wonder how the creator of Fossil feels about AI. Clearly the juggernauts will keep juggernauting but theyre not the only options to collaborate on a codebase. My dream is that more people learn to Just Use Git without the megacorp baggage
Yes, and it's pretty important there's commercial alternatives left on the market.
A few decades ago it would be a different story, but as of now the appeal of going for Github/Gitlab is usually not the git integration and more the rest of it: the issue/MR/PR management, the whole integrated workflow, the CI ecosystem, user and right access management, etc.
I think there must a decent number of organizations keeping a private/local code repository but sync it to github/gitlab to get all the other benefits.
> 4. I do not click or read any articles with AI-generated images or text.
How can you possibly know the articles you read have no AI-generated images or text?
Not OP, but it’s extremely easy to tell if an article uses AI images.
Are there vague images padding the article for SEO purposes?
If so, is there attribution to an artist? If they contain images and don’t attribute, I avoid the website. Either they use AI to make SEO bait, or they steal artwork. I’ll avoid in either case.
AI text is much harder to detect, not sure of a good way to avoid it at the moment.
>If so, is there attribution to an artist? If they contain images and don’t attribute, I avoid the website. Either they use AI to make SEO bait, or they steal artwork. I’ll avoid in either case.
I'm not sure why you think every use of AI generated images is "SEO bait". I'm sure some (most?) are, but it's perfectly plausible a well written article uses AI art in place of a generic image off unsplash or whatever.
> I'm not sure why you think every use of AI generated images is "SEO bait"
Most of the internet is SEO bait. It’s the safe default assumption.
Lack of attribution doesn’t tell you anything. If they’re licensing their images from a stock photo site, attribution likely isn’t required.
Hadn’t thought about that, you make a good point. However in that case the author didn’t feel it necessary to explicitly say the image is not AI generated, which means they likely share very different views about AI artwork and I’ll probably avoid.
Even without that caveat, if they feel it beneficial to pay for stock photos for an article I’m probably good giving it a pass. Most major stock media companies are buying fully into AI generation anyways, and I can’t think of too many cases where stock images really add anything to an article aside from adding to the page size by an order of magnitude.
So what if his criteria has false negatives? AI or not, ignoring noticeable slop will be a net positive.
I thought it was going to be about escape generative AI output online.
Personally, I run a large-ish niche FB group and I haven't seen a threat of bots taking over there or other online spaces. Is there anywhere that people have seen the bots crowding out the people? My guess is Twitter but that place would already/always a sewer.
> hundreds of thousands of Instagram users posted a block of text to their accounts hoping to avoid the plague of artificial intelligence online. [...] “I do not give Meta or anyone else permission to use any of my personal data, profile information or photos.”
Actually, you did. What you didn't do is read the fucking terms of service.
But sarcasm aside, it is a general failure of the system that things have gotten to this point. The amount of naivete like is seen above is tremendous, yet somehow we and everyone who understands this crap have failed to deliver the memo.
"People just submitted it. I don't know why. They 'trust me'. Dumb fucks."
-Mark Zuckerberg.