If AI can dream up a million impressive wood carvings, it seems like the real carving I saw in person is the only one that is actually interesting.
I think whether AI content will put people off the web in general will depend on whether it's actually any good or not. 'That looks like an AI made it' is usually an accusation of shoddiness at the moment!
I guess AIs could always be made to emulate that, but it seems it could make legitimately human stuff have higher value. Sharing with your friend a bootleg live recording that you made might be a greater gift.
I could certainly imagine digitally-signed material from human artists, as well. Maybe I'm just old, but when I'm consuming some kind of art, I want to know that thinking, feeling humans were behind it. If I just want to look at pretty pictures, AI is fine.
And I am actually old, because I'm already put off the modern web. :) HN-style websites are just my speed.
Luster? Huge portions of the Earth depend on the web as their only info source. News has turned into "somebody ran their mouth on (the symbol formerly known as Twitter)."
Also, utterly false news: https://arstechnica.com/ai/2023/12/these-ai-generated-news-a... The comments are disturbing "I always suspected Anderson Cooper and Lisa Ling were computer generated."
Also, you might not be able to read this, because I might already be shadow / hell / heaven banned. (A 2012 update to Hacker News introduced a system of "hellbanning") https://en.wikipedia.org/wiki/Shadow_banning Far too easy to write a script that says if( [user] ) then ( "hide" ).
Also, if you can, we might just live in a ban circle together. Since all news in the future will be personalized based on your tracking pixel and router location. 'ChatGPT, write me a fake nightly news cast for "so and so" location.'
This may be a bit tangential, but something like 1/8th of surviving classical Greek literature (1000 years!) is attributed in part to Galen. While he was quite prolific, the point is that the entire corpus is extremely small.
We didn't even have the capability to record audio and video until the early 1900s, and certainly not in a mass, centralized, easily-distributable sense until the internet. Having a million monkeys write Shakespeare is pure happenstance, but having a million Shakespeares is inevitable. I abstractly worry that even besides the junk, there will actually be so much PHENOMENALLY GOOD stuff that would-be creators and artisans will be discouraged instead of inspired.
Let's hope that you're right about the real carvings. That at least is not so scalable.
Often referred to as the dead internet theory
On the other hand I might be able to print out a copy of the AI generated one, hang it at my desk, and enjoy looking at it a few times before it becomes part of the background of my disposable life.
Compared to a "tribal wood carving designed by GPT" a real world piece of wood is, well, real. The novelty of everything wears off eventually so it won't matter if GPT could still show you nearly infinite carving designs. Without that being turned into actual physical objects by people, they are disposable bytes of data with no real meaning.
In the olden times, you could enjoy an account of a story like that on Facebook because it was supposed to be real (it often wasn't, but there was at least a kernel of truth). Now it seems like I can only experience "authentic" stories / content via first or very limited second hand experience (from a select group whom you trust)
Though if you're an expert in a subject you ask it about, you can see when it transitions from "knowledge of the subject" to "Your uncle who made a living hustling pool and can tell a story made of nothing but whoppers of lies, sounding like the most honest person you've ever met."
It's reasonable to assume it does the same thing over subjects you know less about.
Hopefully soon we'll see it reading the story itself, stripping out irrelevant ads/SEO, and summarizing it.
I actually have a dream of one day going back to a flip phone where I'll get an AI call summarizing emails or messages that are important, as well as any relevant news stories where I can ask for more detail if needed.
The Internet is amazing, but honestly I don't know that the degree to which I personally engage with it is the ideal cost/benefit as opposed to using an intermediary.
Do it. Sonim makes some absolute brick chonks of phones these days.
> ...where I'll get an AI call summarizing emails or messages that are important, as well as any relevant news stories where I can ask for more detail if needed.
I just try to not have unimportant emails come in, though eBay can't be convinced to turn off their damned shipping notifications (or I've not found how to do it). The 1st of the year is a very good time to start wrangling your inbox - every email that comes in is either of value, or you figure out how to eliminate it. Unsubscribe from everything you don't really care about.
And news summary emails are a thing. I route them and various Substack newsletters to my Kobo via Pocket.
> The Internet is amazing, but honestly I don't know that the degree to which I personally engage with it is the ideal cost/benefit as opposed to using an intermediary.
You get a lot of the benefits with far less use than most people assume sane.
I check email via mail clients, when my desktop computers are on. It doesn't come to my phone. I'm off instant messengers most evenings. And I turn my cell phone off at night (which also leads to, on the Sonim XP3+, about two weeks of useful battery life).
Just start.
ChatGPT is essentially an echo chamber between you and ChatGPT.
The point is that if I ask about topics I don't know about, then by definition I can't judge what is or isn't true. That is also the case on HN where I'm reading a discussion about a topic I don't know about, but that on HN there are some signals such as votes and replies to comments, whereas on ChatGPT there are no external signals. Is this perfect? Of course not. Does it usually work kind of okay? I'd say it does.
For books or sites or most things these signals exist too; I can look up what others have said about it.
Even your uncle at a birthday party has some controls because other family members can call him out on hos bullshit, or they can later tell "psst, what Jack told you is a load of rubbish", or whatever.
There is no "generic review" of ChatGPT because it tells every single person someone else. There is no shared content we're both looking at.
ChatGPT very much works in a substantially different way than most other public sources of information.
That's no different from ChatGPT. The control on books at least is a publisher, but that is not really the case for a website. That's why fake news is a real issue.
And your uncle may talk to you in private. Or maybe it's your parents around the family dinner table. There are many sources of dangerous misinformation, not just ChatGPT.
Education and critical thinking is the real solution.
That there are crackpot books out there is besides the point; I can trivially find out which the good books and crackpot books are. With ChatGPT you need to verify every single reply it gives you.
And there is nuance here. Take Horrible Histories. I like Horrible Histories – it's fun, and generally the history is accurate. But it's not perfect and has mistakes, and there's a ton of lists enumerating these mistakes, so you can safely watch the show and then read these lists to ensure you're not believing a mistake.
There is no "historical inaccuracies in ChatGPT" list. It would be impossible to construct such a list. You need to verify every single reply it gives you, and every detail of that not just the general gist.
General critical thinking helps you surprisingly little because many of the signals of reliability that you usually rely on are lacking. Thinking you can trust ChatGPT on topics you're not already fairly knowledgable on and can't be easily empirically tested (e.g. math, programming to some degree) is a good way to get fooled.
Also that information is by definition old. For something it doesn't matter, for many other things, like current news, it very much does.
So while the attraction of having a personal "butler" doing the hard work for you is great, no doubt, it is ultimately no better than getting one's information from Facebook - you have ceded control and get whatever the creators of the system deem you worthy of having. Including political and other agendas. Just look at the censorship the Baidu bot is applying and the entire "jailbreaking" industry around escaping the artificially imposed boundaries on these bots, for good or bad reasons.
And ultimately even ChatGPT and its ilk won't help us any when the web they scrape will be filled with machine-generated spam and crap drowning out all the useful information. Garbage in, garbage out applies even to these robots. They aren't magic.
If you're old enough, similar arguments existed about "open / free internet" and we're seeing some of the consequences of having it be controlled by mega corps.
The LLMs don't have any sort of magical way of filtering that out. So if you train an LLM on spam, hoaxes and similar garbage you will get recommendations to drink bleach as a cure for covid. And that applies regardless of whether the bot is trained by a megacorp or someone in their garage using open source network and tools.
But I don't think there's that much of a watershed date, more like a slow march that accelerated during the Trump era and then went into overdrive after 2020.
Also, algorithmic feeds, "influencers" and "content".
Why should it? 50% of IP traffic is caused by robots, maybe in the future 99% will be. Robot shit viewed by AI robots, SISO. Shit in, shit out.
For instance, how many Reddit comments are bots? With the extreme groupthink present there, it wouldn't be a huge shock if it already makes up over half of the comments in popular subs.
On the popular subs if you read at least half the posts (not comments) you'll see "Op is a bot that reposts content". They typically have some algorithm that digs up old popular posts that haven't been posted in a while and reposted them. With the number of people complaining they are reposted crap I have my doubts their initial upvotes legit, instead getting enough bot votes to get them on the subs front page which then gains more natural votes.
Real question is who pays for 'em?
ePUB books are basically HTML and CSS, so it's a matter of making a "book browser" instead of a web browser, that can inject links to things mentioned in the book text: such as a name, a place, and especially to another book.
Who knows, maybe AI people _want_ the Web to fill with garbage so people have to ask their creations for information.
The shadow libraries make no money by sharing books, instead the librarians risk prison for doing it. So there is no incentive for them to include AI spam books. There is also very little incentive for creators of AI spam books to try to have them included in the shadow libraries.
That's just one incentive. There are others, like ideological battle (e.g. flood the last good information channels with generated myside propaganda, which then will motivated a counter-flood of otherside propaganda, and the mean of the floods isn't truth).
And in any case, ads make a lot of money, but what exactly is an ad anyway? Is there any particular reason an ad can't take the form of an AI-generated spam book in a "shadow library"?
The taint will follow the users.
Would you be able to flood a physical library with AI-generated books?
About two minutes of googling and testing shows that Library Genesis has an anonymous upload function.
I really doubt a pirate library will have the manpower to fully vet every upload. Especially with generative AI able to generate something that sorta looks right.
> Would you be able to flood a physical library with AI-generated books?
A physical library does not have the same characteristics as an online "shadow library," so the question is moot. For instance: a physical has a budget and staff and orders books to shelve, while a "shadow library" must rely on (illicit) donations.
This is the process of LibGen uploads:
> After successful uploading you need to fill the bibliographic data and submit the form, otherwise the file won't be actually added. The file will be accepted to the collection (and become widely distributed) once it gets necessary positive votes from the librarian team, usually it happens in a few days after uploading.
If it's easy to compromise the shadow libraries with AI generated content, I suggest you go ahead and do it for proof.
There are real people running Reddit too (for instance), including mods. That's not an insurmountable barrier.
The shadow libraries have a particular vulnerability to spam, because AI-generated junk can overwhelm their capacity to moderate, but the natural response to the problem (block uploads) could ultimately kill the site. I imagine the main reason they can keep on top of things now is because there's not much interest in subverting their platform, but if it somehow becomes one of the "last good things" on the internet, I think that will change.
Perhaps AI will finally deliver the message.
But then... What if AI dreams the carvings and then drives the CNC? Would that be interesting? Or is there something special about the human being?