Elon Musk & Matt Taibbi Attack the Science of Misinformation

This post contains a video, which you can also view here. To support more videos like this, head to!


It’s (current weekday that you’re watching this) and you know what that means, it’s time for me to talk about the spread of misinformation on social media! I know, I’m sorry, but I really just do not get tired of this subject: social media has provided scientists with a new, large dataset that can allow them to track in real time how people spread lies, either knowingly or not. And for the past 15 years or so that I’ve been following this field of study, the only real “controversy” I can think of is the problem of how ethical it is to study people’s (public) social media posts and collecting their data without their consent, but for the most part.

But a few months ago, a new controversy began to gain steam over the study of social media misinformation: is this research, in fact, STATE CENSORSHIP?? And how did this ridiculous argument gain traction? Elon Musk. Of course. Oh, and his paid army of shills, most notably Matt Taibbi.

“The Twitter Files” is one of those stories I had planned to cover when it first happened in December of last year, but I just got way too busy and decided to skip it. I had no idea that four months later this would still be a thing, and not only that but it’s gotten way weirder and way funnier. But back in December of 2022, Twitter’s new owner Elon Musk released a bunch of internal documents and communications to a few journalists who had been nice to him. The first person to release his “findings” was Matt Taibbi, and of course he released it in a series of dozens of tweets alleging that he was revealing a disturbing trend of the liberal US government working directly with Twitter to censor conservatives. “Because Twitter was and is overwhelmingly staffed by people of one political orientation,” he wrote, “there were more channels, more ways to complain, open to the left (well, Democrats) than the right.”

He went on to describe the way the BIden campaign “censored” the Hunter Biden laptop story, without pointing out that at the time, the campaign was NOT the US government (as this was before the 2020 election), and that the “censorship” simply involved the campaign flagging tweets that linked to graphic photos of Hunter Biden’s genitalia that were released without his consent (because revenge porn is specifically against Twitter’s policies). Considering that the President of the United States at the time asked Twitter to remove a mean Tweet about him from the model Chrissy Teigan, I just can’t really find it in my heart to care about this.

So I didn’t pay much attention to the next few iterations of “THE TWITTER FILES,” for the same reason I didn’t pay attention to the next few iterations of Transformers movies after the first one. Well, not the FIRST one, I mean the first live action one. The first one that didn’t feature Orson Welles.

Yet still, Musk’s friends continued to release more documents over the months, to less and less of an audience. In fact, last month Taibbi released the NINETEENTH Twitter Files thread. NINETEEN! This one was titled 

“The Great Covid-19 Lie Machine

Stanford, the Virality Project, and the Censorship of “True Stories””

Wow, sounds serious! Also it sounds like the previous 18 iterations failed to establish any damning connection between Democratic government officials and censorship on Twitter, so for this round, Taibbi turns his criticism from “government” to “government-funded organizations.”

The Election Integrity Partnership (EIP) was a collaboration formed in July of 2020 between The Stanford Internet Observatory (SIO), The University of Washington’s Center for an Informed Public (CIP), social media analytics company Graphika, and The Atlantic Council’s Digital Forensic Research Lab (DFRLab). Their purpose was to track the spread of misinformation across social media, specifically as it pertained to the 2020 US election, in the hope that society might find better tools to stop falsehoods that, for instance, might lead to a bunch of people storming the US Capitol. To pick a random and unlikely event.

Because as I’ve talked about MANY times in the past, stopping the spread of misinformation is extremely difficult! These institutions have been studying this for years, often finding inconvenient results like “trying to correct misinformation often leads that misinformation to spread further and faster.”

Over the next six months, the EIP partnered with a multitude of entities, including government organizations, social media networks, and journalists, to collect reports of potential misinformation spreading online. They looked at about 22 million examples, and among them they brought about 3,000 to the attention of Twitter to suggest that they violated the site’s own policies.

I think it’s pretty obvious, if you read the EIP’s fascinating and thorough report that is available online for free, that they aren’t the government and they aren’t censoring anyone. They are literally doing the thing these labs are set up to do: evaluating the way that false narratives spread. That’s a good thing! And the fact that they let the social media platforms know when something particularly egregious was happening? That’s also a good thing!

But it doesn’t fit Elon Musk’s narrative, which is that he is a free speech warrior who is here to save Twitter from the evil Democrats. And so it doesn’t fit Matt Taibbi’s narrative either, which may or may not be why Taibbi ended up tweeting some pretty sloppy work, like claiming that EIP reported 22 million posts to Twitter to be censored (when in fact they reported about 3,000 potentially misinformative posts, of which Twitter took action on a fraction of them). Or how he claimed that EIP partnered with CISA, the Cybersecurity and Infrastructure Security Agency (an office of the Department of Homeland Security), and said Twitter didn’t distinguish between EIP and CISA when choosing to “eliminate” “millions of Tweets.” In the screenshot he used, you can see that the email is actually referencing the distinctly non-government nonprofit Center for Internet Security, or CIS, not CISA. And, of course, it wasn’t “millions” of Tweets. And they weren’t “eliminated” – some were removed, others were flagged, and many of them were simply ignored.

These are egregious “mistakes” that, once corrected, make Taibbi’s entire thread collapse under its own weight, but I do want to point out that he also spends a good deal of time whining about EIP’s very open inclusion of NOT just blatant falsehoods but also the other rhetorical tricks that push misinformation, like the use of “facts” framed in a certain way to push a certain narrative. For instance, it may be true that such and such a celebrity died shortly after they were vaccinated, but Tweeting that without mentioning that they, say, were hit by a car IS still contributing to the spread of misinformation. 

And that’s exactly what a LOT of Taibbi’s Tweets (and Bari Weiss, and the other “Twitter Files” writers) do: it’s not ALL blatant lies and “mistakes,” but also picking and choosing what to reveal and what words to use. For instance, complaining in dozens of Tweets about “the government” asking Twitter to “censor” content when you’re really talking about a CAMPAIGN asking Twitter to honor its own rules against revenge porn, with only a brief mention that the ACTUAL PRESIDENT IN OFFICE was asking for mean Tweets to be deleted.

So it is with the utmost irony that EIP members were forced to respond to Matt Taibbi’s thread to try to, well, counter his misinformation by telling people the facts: “As mis- and disinformation researchers, it’s distressing — though perhaps not surprising — to see some of the very dynamics and tactics we study being used to disrupt and undermine our own work and its impact. 

“One thing we have learned is that some of the most effective false narratives work not by spreading outright falsehoods, but by selectively seizing upon and mis-contextualizing bits of factual information, layering those with exaggerations and distortions to create a false impression. Unfortunately, these false impressions aren’t easily refuted through facts that counter individual claims. Often, those rebuttals just provide more ammunition for additional misrepresentations.”

Well, at least they tried.

Another person who tried was Mehdi Hasan, who hosts a TV show on MSNBC. You see, last month Twitter apparently blocked several high-profile accounts at the request of India’s far-right government, to which Hasan commented “I’m sure Taibbi is all over this.” Taibbi replied, “Why don’t you invite me on your show to talk about it? Since you’re so absolutely sure of what I’ll say.”

So, Hasan did have him on his show, at which point Taibbi absolutely shit the bed. In regards to Twitter’s censorship at the behest of the Indian government, Taibbi claimed he didn’t know anything about it and so had nothing to say. But while he was there, Hasan asked about his various inaccuracies, which Taibbi also simply had to admit were mistakes. It was a very enjoyable watch, but it occurred to me that I enjoyed it because I already knew all the things I’ve shared with you in this video today. For the average person, it’s just SO MUCH information that I wonder if it really made an impact. The Twitter Files are really just a microblogged Gish Gallop, in which a person lists off so many half-truths and misrepresentations that their opponent would require an order of magnitude more time and energy to refute.

The amusing coda to all of this is that when Hasan invited Taibbi to criticize Musk for anything, be it censorship of liberal voices, union busting, or what have you, Taibbi refused and stated that he liked Musk. Just a day later, Substack (where Taibbi makes most of his income) launched a microblogging option called Substack Notes, which Elon Musk took as a Twitter competitor and so he made it impossible for anyone to interact with any Tweets linking out to Substack. Taibbi, who previously criticized pre-Musk Twitter for censoring Tweets and allowing prominent people to contact them directly about problems they had, contacted Elon Musk directly to find out why he was being censored. When Musk didn’t fix it immediately for him, Taibbi announced he was going to move from Twitter to Substack Notes, at which point Musk stopped following him on Twitter.

A truly banner week for the Leopard’s Eating Your Face Party. Kudos all around!

Rebecca Watson

Rebecca is a writer, speaker, YouTube personality, and unrepentant science nerd. In addition to founding and continuing to run Skepchick, she hosts Quiz-o-Tron, a monthly science-themed quiz show and podcast that pits comedians against nerds. There is an asteroid named in her honor. Twitter @rebeccawatson Mastodon Instagram @actuallyrebeccawatson TikTok @actuallyrebeccawatson YouTube @rebeccawatson BlueSky

Related Articles

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Back to top button