Elon Musk blames data scraping by AI startups for his new paywalls on reading tweets
Elon Musk blames data scraping by AI startups for his new paywalls on reading tweets
Elon Musk blamed AI companies for new “temporary” limitations on Twitter access. Verified accounts can read up to 6,000 posts per day, while unverified users are limited to seeing 600 posts per day.
He really thinks people want to pay $8 USD a month to scroll through bots, ads, and trolls all day long.
New Reddit can give you the same experience for free. If you want an endless feed of bot garbage, there are competitors doing it for free.
I feel like the limits were chosen to force twitter addicts to upgrade to blue.
6,000/day isn't that much since I heard it counts all replies, too.
I think this might also be related to their Google Cloud bills and inability to move off that platform fast enough. Now they don't want to have to pay for the usage.
Lol without users it's even cheaper.
I've only ever been the low person on the tech company totem pole--but it's certainly an interesting realization that even I could have foreseen many of these issues he's been having these past few days myself.
What's the opposite of imposter syndrome? Because that's when you have the skills but still feel like you shouldn't be there.
What's it called when you know your skills aren't great, but you're also achingly aware you'd do better than that other guy over there?
I hope more and more people wake up to how generational wealth gives people very significant opportunities compared to coming from a family of no wealth/connections.
Being rich is far from a measure of merit in out hyper-capitalistic culture and the disparity between what people deserve and what they get is only going to continue getting worse.
Elon Musk is our new poster boy for absurdly undeserved wealth and there are still far too many who will argue in his defence, from either their position of lucky birth or even from their own delusional poverty.
True! Most of it should have been common business sense; especially after seeing what happened with Reddit, CEOs should know that no platform is permanent.
The Dunning–Kruger effect 🙈
Seems like he shouldn't have bought the bird website and fucked around with it. Not saying I'd do any better if I bought the bird website and fucked around with it, but Jesus Fuck, you'd think you'd buy the bird website, watch it for like a year, and THEN fuck around with it once you've watched it for a while?
The guy has too many doubloons to know what to do with. He should probably slow down or something.
I just don't believe it has to do with site scrapping. He says the daily read cap is temporary measure but that makes no sense. First of all he gives no length of time. Is it days? Weeks? Several months? Scrapping will just resume when the daily cap ends. In the meantime all the users will get frustrated and leave. It's hilariously stupid decision.
It's not a decision I think - I twitter has now tried to move away from AWS/Google Cloud(?) for its hosting but now the bill is due. Elon doesn't want to pay the full billion dollar bill and AWS/Google Cloud(?) has rate limited them because Elon is stiffing them.
So they have to rate limit the users to be able to use the AWS request bottleneck, while implementing that, they DDOSed themselves:
Blaming AI companies as a reason for making stupid decisions, where have I heard this before?
You have to be so candide to believe in this. Scrappers didn’t appeared out of the blue yesterday, if it was to fight it the company would have warn the users about the coming changes.
It’s obvious that’s it’s an emergency change because of him not paying Google Cloud while trying to push for Twitter Blue.
This isn’t how you handle automated processes scraping you, you handle that by restricting things that human beings can’t do. Preventing literally every user from seeing a reasonable amount of content is just cutting off your nose to spite your face.
I’m in IT for a company with a significant web presence. We rely on Akamai for a number of services including CDN and WAF (web application firewall). They have a WAF add-on called Bot Manager that can pretty accurately identify web traffic from bots in real time. If what Musk is claiming was true then tools like Akamais Bot Manager and other similar products would easily identify this traffic and allow them to block it or route it elsewhere.
Musk should know this. I know Tesla uses Akamai.
I'm pretty sure he doesn't wanna pay such service and prefer cheaper / easier way.
This man should've paid his bills and paid Google. Twitter didn't have these problems before Elon.
Of course the manchild with fragile ego blames other people for his problems.
It‘s like a toddler who cries when someone else wants to play with their toys.
I wonder if this is what just automatically happens to some rich people, they never were taught to or needed to share anything, certainly would explain the pathologic wealth hoarding.
I think people are missing the bigger picture here. We're approaching a point where AI might not be feasible because of privacy concerns and copyright infringement. Midjourney is scraping art from millions of artists without their permission or knowledge.
If I look at 12 van goghs (or any artist) painting and tried to make something in a similar style do i need permission? I don't see why how the AI learns is relevent. As long as the creation in the end is original.
Good point, most artists don't live in a vacuum these days. The art we create results from a similar traning process than the AI's. Maybe just treat AI art the same with all rules about plagiarism so that it has to be original as well?
edit: a word
You are not allowed to use any media commercially for free, why would that be different for an AI?
Well the difference between you and an AI is that you have biases. You have what is called a perception filter that everything you've experienced is fed through. You can consume as many Van Gogh's as you wish but you'll never match Van Gogh because you aren't Van Gogh. An AI doesn't have a perception filter, it sees things as they are and will replicate it as best as it can. It'll go to the extent of copying Van Gogh's paintings entirely and someone can take those and post them as new things. An AI cannot be inspired by Van Gogh, it isn't capable of that, it just copies.
People like him never admit to doing anything wrong. If only more people'd learn this sort of behaviour should give them major red flags.
In other words: Man who uses his users' data for money is mad when other people use his users' data for money.