Wikipedia survives while the rest of the internet breaks

false equivalence https://en.m.wikipedia.org/wiki/False_equivalence

Maybe give some arguments or evidence instead of linking wikipedia.

Everyone clicking on that archive link avoided having to pay someone to see information.

I'm personally a copyright abolitionist, so I'm pretty okay with that - but I want people who pearl clutch about AI systems not paying to realize that it's the same damn thing going on when they click on the archive link.

I'd disagree that it's the same thing.

With AI content scraping, the people whose content was ingested aren't getting paid. That part does align to paywalls.

But there's more in the AI content situation. The scraped content is repackaged without any credit being given to the people who made the content. In most cases, the models trained on the content are intended to be monetized, and there is no intent to share revenue with the people who made the content.

When I bypass a paywall, there is no particular expectation that I'm going to take the content, modify it, and display / sell it as my own. In the vast majority of cases, someone reads some free content and moves on. The damage to the site or publication is limited to the unpaid viewing.

AI content scraping absolutely comes with an expectation that the content will be modified, presented, and sold. The damage goes beyond the unpaid viewing of the content.

The submission is not about LLMs being fed copyrighted material during training.

The poster does not immediately seem to be an advocate condemning the practice.

Your post was out of place.

We detached this comment from https://news.ycombinator.com/item?id=45128461 and marked it off topic.