Nobody talking about how this article seems very strongly AI-written? Like it offers things it can "do for you" at the end of each section. And the "what this means to you' has nothing about what this means to you......
Very interesting.
this post was submitted on 22 Aug 2026
197 points (97.6% liked)
Technology
87450 readers
3650 users here now
This is a most excellent place for technology news and articles.
Our Rules
- Follow the lemmy.world rules.
- Only tech related news or articles.
- Be excellent to each other!
- Mod approved content bots can post up to 10 articles per day.
- Threads asking for personal tech support may be deleted.
- Politics threads may be removed.
- No memes allowed as posts, OK to post as comments.
- Only approved bots from the list below, this includes using AI responses and summaries. To ask if your bot can be added please contact a mod.
- Check for duplicates before posting, duplicates may be removed
- Accounts 7 days and younger will have their posts automatically removed.
Approved Bots
founded 3 years ago
MODERATORS
Bye reddit, you were ok, 10 years ago.
Since when does ChatGPT provide citations?
It's had web search ability for about 2 years now. That makes it a fair bit more reliable than before except for when the source it finds is actually irrelevant. But if you're knowledgeable in the domain you're prompting about, it'll usually be easy to figure out. If you're not... Well, good luck.
I know it had web searchability, but it never provided the source of the information, which is, like, a giant fucking problem .
All major LLMs have had citation ability for while now. Any time I ask one a question I can't find the answer to, if it's not providing a source, I automatically assume the output is slop. And even then, you still have to be wary, becsuse it will often make something up then cite a semi-related source that doesn't actually contain its claims.
i cant get any LLMs to consult reddit at this point.
Gemini seems to
The drop correlates with the time reddit locked down old reddit from accessing anonymously/without logging in to avoid AI scraping. Reddit themselves did this.
How much you wanna bet this lazy reporting from "promptwatch" is AI generated?
Even the normal Reddit site locks up after a few seconds if you're not logged in. I'm still happy to kill time on msoutlookit while I'm at work, but follow any links to the main site and it locks you out pretty quickly. I imagine even that will be gone too soon.
only on mobile
No it's happening to me on pc
OpenAI would not have been scraping old ui?
Also most reddit content has long since been injested. No need to visit your http site anymore
They actually did use old reddit, as confirmed by reddit itself, as confirmed by massive drop after the change. It was simply far easier to scrape old reddit (pre-rendered instead of relying on JavaScript and dynamic content load)
You seem to have misunderstood, and the article you linked doesn't contradict what I've said.
Wealthy, sophisticated LLM developers would not have been scraping old reddit. They would pay reddit for API access.
Shutting down old reddit closes the door on back yard developers, not OpenAI.
Additionally, when you run a prompt in a chatbot, it doesn't scurry away and scrape old reddit and then formulate an answer. The scraping of content is going on while the model is being developed.
As if wealthy people pay for stuff they could have for free. Literally everybody had their copyright infringed from the training of GenAI. Authors, Journalists, Artists, Musicians, Regisseurs, the list is endless. Unless you sue, you won't see any compensation from them. By now you should know that companies only play fair if they're forced to, otherwise literally anything goes.
It's quite naive to believe they would pay for API access, tbh.
Wealthy, sophisticated LLM developers would not have been scraping old reddit. They would pay reddit for API access.
Shutting down old reddit closes the door on back yard developers, not OpenAI.
And yet it did. And yet reddit mentioned shutting down API too
Surprised this fucking idiot used a source instead of strategically editing it out.
Imagine if the AI companies used their infinite money glitch to just replace Reddit instead. Now that's karma.
[chuckles in fediverse]
While I'm not a fan of scrapping you would think reddit would want this to push traffic to the site. Get a good answer and follow the referral link and you have a new monkey.
Are there any statistics on how many LLM chatbot users actually click on the source links?
That's just for Google's AI summaries, but it's the only one with actual methodology disclosed I could find on a quick (non-AI) search.
Plenty of SEO mills with their own "studies" too, and the trend is similar. People don't interact with the source sites or if they do it's extremely seldom.
They (SEO mills) are currently trying to flip the gaslight towards "Worry not dear Webmaster, now you just have to make sure your content is now formed so it's engaging for AI search and engage able right from the AI!" type shit.
I'm tired boss, and I'm not even generally (as in, in all aspects) anti-AI. Yet, I guess.
Ah yes, I've noticed lots of SEO consultant companies are trying to pivot into "we'll help you to make AI scrape your website more so you show up in the AI search".
I hadn't seen the study before, thanks. That's what I expected, honestly. I guess you could argue they are still building brand awareness, because people subconsciously note the brand as the source, but I don't think that amounts to much.
I mean just about every metric I've seen suggests most websites that see an increase in scraping see a decrease in actual human visitors. I doubt there's much distinction between chat and search users - the goal of these services is to drop click-through rates.
Oh, definitely. They are reinventing the old portals concept. That's why I don't think Reddit actually wants to encourage the scraping (partly also because they sell data themselves to build LLM, but that's another thing)
Great question, I wish I knew. I know when I'm on a deep dive I will hit the links as there is other information or other articles the scrapped site has.
I don't know how many do, but they're fools if they don't.
You'd think that since those links are there, that the information on those links matches what that LLM says, and thus not a hallucination, but at least twice, I've found the information straight up wasn't there, or the LLM had misread / misinterpreted the page's content.
But then I may not be a regular user. I ask a question maybe once every one to two weeks because a regular search finds nothing and, like I say, I prefer to check the sources.
I do for things where I want to be accurate, but sometimes it seems like I’m the only one
On the other hand I advised my teen to buy the wrong battery because I didn’t click into the results for “what battery do i need to replace in my Subaru key?”
I think it had the opposite effect. I believe the recently didn't renew their contract with Google because they were upset people weren't coming to the site, just getting answers from Gemini. I could be wrong, but I thought I read that
The guys on LTT mentioned this during their Linux challenge, but I’ve found it to be true -
Scouring forums for answers to obscure errors used to take hours. Now I just throw it at Gemini, troubleshoot in real time and find the fix.
I’ve even started asking it to give me a PDF of the steps we took and filing them away for future me.
I like to ask it for a beautiful html page instead of a pdf. PDFs are a pain later on.
Tonight I spent the time spinning up a silverbullet instance and converting the PDFs to .md files. You’re right, PDFs aren’t searchable and will be a pain as the database grows.
Thanks for sharing :)