X updates its terms to ban crawling and scraping

Dupe: https://news.ycombinator.com/item?id=37433759

Y'know, if tech companies think crawling the entire Internet to train text and image generators is a bad thing then maybe they should stop also trying to train text and image generators on Other People's Content. Alternatively if AI is just that good then anyone should be allowed to scrape and train and the resulting models should be public. If AI companies don't think they should pay for training data then I shouldn't have to pay for their models.

As it currently stands Xitter[0] is trying to have their cake and eat it, too. Reddit as well. The idiots running these companies are burning everything to the ground for the sake of trying to build a new monopoly around training data rights that largely don't exist.

[0] Pronounced "shitter"

Warning for those considering this: I got banned from Reddit for scraping it. I was nice enough to include my username in the user agent (I foolishly thought they would be nice and give me a warning instead of straight up banning my 10+ year old account if they didn't like what I was doing).

So when companies do this I genuinely wonder: * How will they know? * What will they do?

They've literally spent years convincing the world that the data on their platform is freely open to the public therefore it's ok if they collect as much data per person that they can.

When I read the title I thought "crawling and scraping...no problem, I can do crawling separate from scraping".

But alas, the terms actually ban crawling or scraping.

Good luck with that.

This is so confusing, every time I see a headline about Twitter I think to myself "what does X.Org have to do with this?"

INAL but wasn't scrapping legal? They can ban it in their T&C but that doesn't make it illegal.