I publish my content "for free" because I want to spread knowledge and information, or promote a topic or interest I enjoy, and monetization has never been a priority for me. I know that I am not alone. The urge to create does not depend on a need for money. I am happy for my websites to be picked up by Perplexity or ChatGPT because I want more people to see/learn/hear about the things I care about.

If someone only creates for money, only publishes on the web to get people to look at advertisements, well... I think there are plenty of other people who don't feel that way that will fill the void left behind in their departure.

To me it seems weird so many people think the internet only exists because advertising props it up. The internet existed and was a wonderful place before advertising became widespread, and most services and websites will continue to exist after advertising is gone (if that ever happens). What encourages people to believe in some sort of great collapse?

I said "or any other reasons" because I was in a hurry. Internet existed way before ads, I agree, but even then you wanted people to visit your site to see what you wrote. Maybe to become an expert in a topic, maybe to feel better, but you want people to know who did it. That's why websites used to have a webmaster and info about that particular person.

If people stop visiting websites because LLM give them what they want, websites will stop existing. Don't believe me? Check how many "fansites" exist now about topics compared to ten years ago, when there weren't social networks. They have been replaced by influecners with huge followers on Instagram, TikTok, Twitter and more. The same will happen.

What you describe in your last paragraph is still an evolution, not an extinction or disappearance. Fan/stan culture is absolutely huge on social media, where it has thrived, with single-person-ran stan accounts becoming as big or bigger than the heyday of blogging or fansites. So I just think we disagree on some foundational level.

Absolutely this! Actually the internet was driven by people who wanted to share their ideas, not by those who wanted to sell their ideas.

I've been thinking about this. There's three main solutions (https://corporatefinanceinstitute.com/resources/economics/fr...) to the Free Rider Problem, which comes up with non-scarce goods like this.

The first is taxation. Academia uses this. So we'll be fine there.

The second is donations and altruism. This includes Wikipedia, people who just want to share their ideas (like I'm doing now), Stack Exchange, etc. So we'll be fine there too, though this only works for low-budget stuff. Note that credit isn't necessary here; Wikipedia editors are rarely acknowledged, for example. But I do think credit is good, when possible (i.e. not for every single training source the LLM uses; ChatGPT Search is a good and practical way to give credit).

The third is to privatize the goods, for example, by creating copyright. LLMs won't get rid of this. They're not allowed to fully replicate a piece of text, since that is copyright infringement. So if you're getting people to pay for an exact or almost exact piece of text (e.g. a full book, a full newspaper article), you won't be affected there.

But if your business model would be affected by people summarizing your stuff, then yeah you won't be protected there. For example, if people would rather read a summary of your article or book rather than pay you for the full version. But this isn't new to LLMs. How many people jump straight to the comments to read the summaries instead of reading the article? How many people absolutely refuse to pay for paywalls? How many people block ads (I'm treating ads as a form of payment here)? It's possible to expand copyright to also cover things like summarization (i.e. copyrighting facts), but this is rather dangerous.

Seems like a good thing to me.

- The clickbait, SEO-optimized garbage that today fills 95% of search results could entirely disappear as a business model because they have nothing interesting to offer and the LLM company won't pay for low quality content.

- The average Joe blogging on their website won't go anywhere because they aren't profiting from it to begin with. And the LLM linking back to the page with a reference would be a nice touch. Same logic applies to things like Open Libra and projects that are fundamentally about open information and not about driving ad revenue.

But, on the other hand, I don't think LLM-based search will fundamentally change anything. Ad revenue will get in the way as always and the LLM-based search will start injecting advertisements in its results. How other companies manage to advertise on this new platform will be figured out. What LLM-based search does is give Microsoft and others the opportunity to take down Google as the canonical search engine. A paradigm shift, but not one that benefits the end user.

> A paradigm shift, but not one that benefits the end user.

That's too neutral; the result is worse for the end user. It will be impossible to distinguish injected ads, unlike with Google. Furthermore, as it gives a much more direct "answer", people will also trust it more than a website linked to on Google.

Yeah, I was being generous in the argument :)

Well the whole point of this product is to link back to websites. There’s no necessary link between the text and the links, which are chosen after the fact from an index. That’s different from traditional search engines, where links are directly retrieved from the index as part of ranking.

The answer though has so much info already that it reduces the need to visit the links.

Honestly, for any serious query, the links serve mostly so you can double-check the AI. That's a useful function however.

I think we're going to see even fewer site visits as a consequence of AI search engines. The internet's ad-based funding model is going to dry up further, but the impact will be disproportionate. It'll be a few years til we see where the cards land.

It would make sense for peoppe to be able to "connect" things like the newspaper they subscribe to to their LLM agent. The LLM would essentiay be a new frontend.