The Science of Censorship: AI Censorship Models and Elon Musk’s Counter Move

Elon Musk has applied temporary reading limits to clamp down on “extreme levels” of data scraping and system manipulation.  In a tweet, he outlined that verified accounts have been limited to reading 6,000 posts a day, while unverified accounts can read 600 a day. Newly unverified Twitter accounts will only be able to read 300 posts each day. In a subsequent tweet, he wrote: “Rate limits increasing soon to 8000 for verified, 800 for unverified & 400 for new unverified.”

The Tesla and SpaceX CEO said Twitter had imposed the “temporary limit” to “address extreme levels of data scraping & system manipulation”.

The disninformationists, free speech opponents and media propagandists are decrying Elon Musk’s new rate limit policy as some sort of censorship (laughable coming the worst oppressors of free speech). But from the lens of the censorship industry Elon Musk is actually “preserving the openness of the internet by preventing the construction of this censorship Deathstar that is getting more and more refined every day and it’s being funded by your tax dollars to the tune of tens of millions from DARPA and the National Science Foundation to say nothing about the State Department and USAID and National Endowment for Democracy grants…’

You could not censor the internet just a few years back at the scale you can now because we now have AI censorship models. It all relies on massive scraping of Twitter date in order build these models and data bases to track trending narratives to systematically surveil to track and turn down all at once communities online. Mass censorship on an unimaginable scale

Mike Bentz Executive Director, Foundation for Freedom Online and former State Dept cyber explains, “Elon Musk has no idea the DARPA rattlesnake he just stepped on by this doing this…”

Watch this.

Wow. Musk has no idea the DARPA rattlesnake he just stepped on by this doing this…

My take on @elonmusk’s new rate limit policy, from the lens of the censorship industry: https://t.co/AZ5SPIgHiz pic.twitter.com/Rw4KBZsT9O

— Mike Benz (@MikeBenzCyber) July 1, 2023

 

https://twitter.com/KanekoaTheGreat/status/1675330302296854529?s=20

Excerpt:

 I’m going to sort of address this issue from the perspective of the censorship industry because it’s a really fascinating thing that Musk just did with with wide ranging implications for the science of censorship.

…AI censorship is where all of the magic happens. The Twitter files showed how the FBI might come in and get 22 tweets censored, but AI was how EIP (Election Integrity Partnership) and other type of third party censorship groups were able to get 22 million tweets, censored so that’s six orders of magnitude, right? It’s a completely different animal.

You could not censor the Internet before 2016 at the kind of scale that you do now because you have AI censorship models. There’s a AI censorship deathstar that has been under construction and innovation and renovation every every week, every month every year for the past six, seven years now and it all relies on massive scraping of Twitter data in order to build these models and databases, to track trending narratives, to systematically surveil and build sort of intelligence dossiers and to track and to turn down all at once, communities online (political communities, social or public health communities, you know, any climate communities, you name it, whatever the sensitive policy issue of the day is.)

You can use this massive Twitter scraping capacity to ingest everyone’s tweets, and then disambiguate out all of the words that they’re using the hashtags, the themes, the memes, and to build this sort of codebook of online communities that can then be used for mass censorship that is used for mass censorship.

So a good example of this is how did EIP end up getting 22 million tweets classified as information? Well, first, they didn’t flag 22 million by hand. This was something that Mehdi Hassan, the MSNBC host, flagrantly deceived his millions of viewers about they when he said, ‘oh, they only flagged 3000 not 22 million, hardy hardy hary are so you only off by 21 point 7 million…..’

No, you butthead actually what happened was is they used AI to designate 22 million because they scraped 159 million tweets to build their AI model of which they said 22 million constituted misinformation. The 3000 was just the tip of the 22 million iceberg because back in the Twitter iteration before today, these third party censorship consortiums could just mass ingest hundreds of millions of tweets all at once, in order to build these vast censorship models in order to control all of public conversation all at once.

Now at that time they also had captured the trust and safety teams at Twitter. So it’s very easy for them to build that model. And then, you know, basically target those 22 million tweets and all the, hundreds of thousands of accounts, and have those be automatically throttled because they were controlling the actual enforcement arms as well. It’s unclear if that continues to be the case.

The fact is, is that Twitter is by far and away, the intelligence agency, the State Department, the Defense Department and the NGO Plex social media platform, par excellence. They much prefer the Intel they can scrape from Twitter than from any other social media company. And there’s good reason for this, I guess other than LinkedIn, I’d say and they would say, and part of that is because of Twitter’s not a walled garden like Facebook. Facebook is also a social media platform obviously, but it’s most private accounts you can only access if you are friends with that person or friends on a friend. Twitter does did not have that. And YouTube is a there’s a giant divide between producers and consumers. Consumers on on YouTube are not create content creators. That’s not the case on Twitter. Every Twitter user is a content creator when they click the retweet button. And so you can you can map real time narrative emergence on Twitter in a way that you can’t do on any other social media platform and that has been used by the CIA, the State Department the Defense Department, hundreds of censorship government funded NGOs, centers, nonprofits foundations, university quote unquote research centers and and the works it’s in order to build their social media censorship Deathstar. They need to be able to scrape hundreds of millions of tweets. And what Musk is doing here is very interesting because on the one hand, you would argue you could argue it’s a sort of kind of form of, I wouldn’t say censorship necessarily, because it’s not like a particular person is singled out but you are, you are limiting the openness of the internet by doing this. But on the other hand, you are actually potentially preserving the openness of the internet by preventing the construction of this censorship Deathstar that is getting more and more refined every day and it’s being funded by your tax dollars to the tune of 10s of millions from DARPA and the National Science Foundation to say nothing about the State Department and USAID and National Endowment for Democracy grants. So it’s all very interesting……

Elon Musk claims Twitter login requirement temporary measure due to data scraping

Musk calls out one artificial intelligence leader in particular

By Julia Musto FOX Business, July 2, 2023:

Elon Musk ‘doesn’t care’ if people don’t like what he says: Mike Huckabee

Former Arkansas Gov. Mike Huckabee discusses Elon Musk saying that suppressing the Hunter Biden laptop story was election interference on ‘The Bottom Line.’

Elon Musk assured Twitter users Friday that a requirement to have an account to view tweets was a “temporary emergency measure.”

Users who try to view content on the social media platform are asked to sign up for an account or to log into an existing account.

“We were getting data pillaged so much that it was degrading service for normal users!” the billionaire wrote.

The cause of the change is hundreds of organizations that were scraping Twitter data “extremely aggressively,” impacting user experience.

BIDEN FTC CHAIR PROBING ELON MUSK’S TWITTER PURCHASE TO FACE GRILLING FROM JIM JORDAN’S JUDICIARY COMMITTEE
Tesla CEO Elon Musk

Elon Musk, chief executive officer of Tesla, during the EEI 2023 event in Austin, Texas, on Tuesday, June 13, 2023. (Photographer: Jordan Vonderhaar/Bloomberg via Getty Images / Getty Images)

“We absolutely will take legal action against those who stole our data & look forward seeing them in court, which is (optimistically) 2 to 3 years from now,” Musk pledged.

Artificial intelligence companies are to blame, the SpaceX founder asserted in a reply to one user.

“Almost every company doing AI, from startups to some of the biggest corporations on Earth, was scraping vast amounts of data. It is rather galling to have to bring large numbers of servers online on an emergency basis just to facilitate some AI startup’s outrageous valuation,” he said.

The Truth Must be Told

Your contribution supports independent journalism

Please take a moment to consider this. Now, more than ever, people are reading Geller Report for news they won't get anywhere else. But advertising revenues have all but disappeared. Google Adsense is the online advertising monopoly and they have banned us. Social media giants like Facebook and Twitter have blocked and shadow-banned our accounts. But we won't put up a paywall. Because never has the free world needed independent journalism more.

Everyone who reads our reporting knows the Geller Report covers the news the media won't. We cannot do our ground-breaking report without your support. We must continue to report on the global jihad and the left's war on freedom. Our readers’ contributions make that possible.

Geller Report's independent, investigative journalism takes a lot of time, money and hard work to produce. But we do it because we believe our work is critical in the fight for freedom and because it is your fight, too.

Please contribute here.

or

Make a monthly commitment to support The Geller Report – choose the option that suits you best.

• $25.00 here
• $50.00 here
• $75.00 here
• $100.00 here
• $250.00 here

5 1 vote

Article Rating

4 Comments

Oldest

Newest Most Voted

Inline Feedbacks

View all comments

Zeus

10 months ago

The thing they fear the most is overwhelming numbers of people with their minds set on dealing with the ruling class by whatever means necessary. The more We The People learn how diabolical and hostile and deadly the ruling class really is, the closer we are to getting our minds set on the solution. The closer we get to realizing that our only possible defense is offense, the uglier and deadlier they become. The uglier and deadlier they become, the more we see what we are really up against. At some point, the herd turns to a brutal offense against the lions, as the only possible defense, and when that happens the lions have no chance at all. The “invisible” good guys can attack with the herd.

4

Snowedin

10 months ago

Any way one looks at it AI is dangerous to the world and society and needs to be closely monitored, or it will destroy us.

4

George

10 months ago

Big tech and government censorship is diabolical. They are traitors and should be hung from the the nearest tree. Bastards.

3

John Acord

10 months ago

I hope people are beginning to undersand what a gift Elon Musjk is to the fuure of humanity.

-3