Why isn’t Twitter working? How Elon Musk finally busted his site and why the internet could be getting worse

Musk tweeted this week that the term cis is an 'insult' (Alamy/PA)

Musk tweeted this week that the term cis is an insult (Alamy/PA)

It started out like any other outage: unexplained error messages telling users they’d hit their speed limit and Twitter posts refusing to load. But as the weekend wore on, it became clear that these weren’t just any technical issues, but rather problems that could define the future of not just Twitter but the internet as well.

Elon Musk took to Twitter on Saturday and announced he would roll out a series of changes to address extreme levels of data scraping [and] system manipulation. Users would only be able to see a limited number of posts, and those not logged in would not be able to see the site.

That decision triggered those error messages, as users were hitting the speed limit, which meant they were requesting too many posts for Twitter to handle. The seemingly temporary new limits, though still in effect, meant that users were rationed on how many tweets they were able to see, and would see frustrating and inexplicable messages when they actually hit that limit.

In many ways it was yet another baffling and troubling decision by Musk, whose running of Twitter has gone from scandal to scandal since he took over the company in October last year. (He appointed a CEO, Linda Yaccarino, last month, but she appears to still be deciding, executing and communicating the company’s strategy.)

But this time something feels different in the chaos. For one thing, it’s not one of the many content-policy issues or potentially hostile ways to encourage people to subscribe to Twitter’s premium service that have so far marked Mr. Musk’s leadership of Twitter; on the other hand, it seemed to be part of a larger problem that is shaking the entire Internet and of which Twitter could only be a symptom.

It’s unclear whether Musk’s latest decision really has anything to do with scraping by AI systems, as he claimed. But the explanation certainly makes sense: AI systems require vast corpuses of text and images to train on, and the companies that make them have generated them by scraping and regurgitating text that can easily be found on the web.

Every time someone wants to load a web page, their computer makes a request to that company’s servers, which then provide data that can be reconstructed about the user’s web browser. If you want to load Elon Musks’ Twitter account, for example, point your browser to the relevant address and it will display his Twitter posts, pulled from the Internet.

That comes with costs, of course, including the price of running those servers and the bandwidth required to send large amounts of data quickly over the Internet. For the most part on the modern internet, this cost has also been covered by sending advertisements or requiring people to subscribe to see the requested content.

The AI ​​companies that are scraping those sites, however, are requesting that data frequently and quickly. And because the system is automated, they aren’t able to watch ads or pay for subscriptions, meaning companies don’t get paid for the content they provide.

This problem seems to be growing on the internet. Companies that host text-based discussions, such as Twitter, are very aware that they may be providing the same data that could one day make them obsolete, and they want to at least make some money out of this process.

It also appears to be one of the reasons behind the recent fallout on Reddit. That site is especially good for feeding an AI, it includes very human and very helpful answers to the kind of questions users might ask an AI system, and the company is well aware that, again, he is giving up information that could also be used to overcome him.

To try to fix it, it recently announced it would charge large amounts for access to its API, which serves as an interface through which automated systems can fetch that data. It was at least partially intended as a way to generate money from those AI companies, though it also had the effect of making it too expensive for third-party Reddit customers who also rely on that API to keep going, and those more popular they have since it closed.

There are good reasons to think this will continue to happen. The web is increasingly being sucked up by the same AI systems that will eventually be used to further degrade the user experience: Twitter is, in effect, being used to train the very bots that will one day post misleading and annoying messages everywhere Twitter.

Every website hosting text, images or video could face the same problems, as AI companies try to create their own datasets and train their own systems. Thus, the entire Internet may become more like Mr Musks’ Twitter over the weekend: actively hostile to real users, as he attempts to keep fake users away.

But equally likely is that this is Mr. Musk’s explanation for why the site went down suitably in tune with the zeitgeist and usefully shifts the blame to AI companies about which he’s already expressed significant skepticism. The truth could be that Twitter, which has laid off the vast majority of its staff, including those on its engineering teams, may finally be running into infrastructure issues that occur when fewer people are around to keep the site online.

Twitter’s former head of trust and security, Yoel Roth, is perhaps best qualified to suggest that this is the case. He said Mr Musk’s argument for the new limits fails the sniff test and instead suggested it was the result of someone mistakenly breaking the speed limiter and then that the incident was passed off by Mr Musk as intentional, whether he knows it or not. .

For anyone keeping track, this isn’t even the first time they’ve completely busted the site by fiddling with the speed limiter, Mr. Roth wrote on Twitter rival Bluesky. There’s a reason the limiter used to be one of the more locked-in internal tools. Driving around with speed limits is probably the easiest way to break Twitter.

Mr. Roth also said that Twitter has long been aware it was being canceled and that he was OK with it. He called it Twitter’s open secret of data access and said the company considered it okay.

And he, too, hinted that the weekend’s events could be a clue to what’s coming to the internet, offering an entirely different alternative. It’s not Twitter, Reddit and other companies that should really be upset about what’s going on, he suggested.

There’s some legitimacy in Twitter and Reddit being angry at AI companies for drinking social data for free in order to train commercially viable models, Roth said.

But they should never forget that *their* data is not ours. A solution to parasitic AI must be user-centric, not profit-centric.

#isnt #Twitter #working #Elon #Musk #finally #busted #site #internet #worse
Image Source : uk.finance.yahoo.com

Leave a Comment