- cross-posted to:
- hackernews
- cross-posted to:
- hackernews
Around the beginning of last year, Matthew Prince started receiving worried calls from the bosses of big media companies. They told Mr Prince, whose firm, Cloudflare, provides security infrastructure to about a fifth of the web, that they faced a grave new online threat. “I said, ‘What, is it the North Koreans?’,” he recalls. “And they said, ‘No. It’s AI’.”
Those executives had spotted the early signs of a trend that has since become clear: artificial intelligence is transforming the way that people navigate the web. As users pose their queries to chatbots rather than conventional search engines, they are given answers, rather than links to follow. The result is that “content” publishers, from news providers and online forums to reference sites such as Wikipedia, are seeing alarming drops in their traffic.
As AI changes how people browse, it is altering the economic bargain at the heart of the internet. Human traffic has long been monetised using online advertising; now that traffic is drying up. Content producers are urgently trying to find new ways to make AI companies pay them for information. If they cannot, the open web may evolve into something very different.
Archive : https://archive.ph/nhrYS
why do you say it’s not killing the web? How many people are on lemmy because of the enshitiffication of privately owned social media?
Because this is what recipe sites used to look like, which was a person, posting recipes they had, in their words, so that you could get the information. Could the site look a bit better? Sure, there are accessibility issues, but it’s honest, to the point and gives you what you’re looking for.
This is the first result on Duckduckgo for “Garlic bread recipe”, which contains umpteen paragraphs of irrelevant information about Tony, a ton of great pictures of garlic bread, but holy fucking shit, they’re both the same content, one is just insufferable and ridiculous and the other one gives you what you’re looking for.
I’ve been in web design for as long as the web’s been around, basically. Obviously what would be nice is a better laid out site that gives access to some of the beautiful pictures that are in the 2nd article with the content of the 1st. But you can’t write an article with as few words in the first and ever hope to be ranked above another with “more content”. SEO has driven the web to extrude out content for attention that can be better dragged across advertisements.
For someone just looking to quickly understand how to make a good garlic bread, I miss the old web dearly.
re: enshittification of social media, social media isn’t the world wide web, so I am not understanding how that relates, but that said, I dearly also miss the forum-based internet and IRC as primary ways of communicating with each other. You were a lot less likely to run into all the giant personality conflicts that happen on Reddit/Lemmy because you weren’t aggregated in with literally everyone to comment on literally everything, you were organized around niche interests.
So I guess my question back to you would be, how many of us are on Lemmy/Reddit because that’s all that’s practically left for us?
as I understand it, you’re saying the web was attacked by SEO, and that sounds entirely correct, but I don’t follow how that means AI isn’t also killing the web.
tbh I don’t have the energy to contribute meaningfully to this conversation so I’ll stop before I say something silly
The two things are related is what I am saying. The SEO web resulted in inflating pages with nonsense content that is needed to better appeal for advertisement revenue. This results in the results of web searches being clogged with in-genuine content that is meant to appeal to google’s algorithm and not to/by/from actual people.
The idea of “the web is dying” would mean that the internet has gotten so clogged with this kind of junk that no one even bothers to use it anymore. That’s not what’s happening. Instead, for better or worseª, people are getting parsed LLM search results at the top of their query which is summarizing the “content-rich” articles that are built for google, and getting people to the answers they want faster.
What that’s doing is making the idea of putting together this elaborate peacock of a website to appeal to Google a lot of wasted effort, since no one is going to those pages if they can get the answer more quickly. Whether this sounds like a bad thing to you I suppose depends on how much you value the pre-AI but post-SEO period of the internet, which I guess I don’t have much nostalgia for.
FWIW I’ve gotten some fantastic help from search engine LLMs with respect to some broad coding tasks but i have always said i don’t think it’s taken any longer than it would have for me to go to StackOverflow and parse through answers on my own, especially after tweaking and working with the LLM to refine its answer. I don’t think LLMs are magic, I think they’re often slapped into poorly designed products and overly applied but can have some limited use.
ª (Often worse. I got an unasked-for LLM search result on DDG today for ‘What is my IP address’, a question an LLM could not possibly help or answer without being hooked into a lot of other things, and is just a colossal waste of time and energy, and is really bad design)
Because what is dying is one part of the web, or more specifically, a certain type of product that is served over the web.
Which I, for one, am very, VERY glad is dying. There’s an infestation of SEO driven sites with basically a lot of word diarrhea and metric tons of ads, and finally at the bottom, the single sentence you were looking for.
A pox on those sites, let them die.
it would be nice if you’re right. I don’t know enough to comment. But I don’t have confidence in market forces, for very good reason. My concern is that bad will be replaced by worse