arctotherium Profile picture
Mar 24 22 tweets 10 min read Read on X
To make fact-checking work during the closure of the Internet, social media platforms had to know the ground truth of claims. Since this is not precisely knowable, they outsourced determining the truth to a web of news organizations and NGOs. Thread on these. Image
Most official fact-checking organizations were certified by other the International Fact-Checking Network (IFCN) or the European Fact-Checking Standards Network (EFCSN), which created a chokepoint in the ecosystem. Image
The IFCN was founded by the Poynter Institute, a school of journalism, in 2015, after a $1M foundation grant. They hired an ex-SPLC employee to create a list of 515 orgs to be used in ad blacklists, including mainstream conservative ones like the Washington Examiner. Image
Image
Sometimes, claims are just wrong. But often, they are judged by fact-checkers as misleading for missing context (which may or may not be true - but what context is relevant is not prima facie obvious) or partly false based on nitpicking. Both types get suppressed. Image
Meta's initial source of ground truth was five US fact checkers, hired in 2016: PolitiFact, Factcheck[.]org, Snopes, ABC News, and the AP. All are correctly known to be strongly leftist and Democratic orgs, albeit with more neutrality than one like HuffPo. Image
Another IFCN-like "alliance of news media, social media, and technology corporations" to "combat disinformation" was the Trusted News Initiative, founded in 2019 and closely linked to the British government. Image
An example of one of the most academic orgs used by these various alliances was the SIO, the Stanford Internet Observatory, funded by the US government. This was eventually dismantled after House Republicans noticed it kept id'ing conservative arguments as disinformation. Image
Image
Another similar org is the Digital Forensic Research Lab (note the date: founded 2016). Image
Hamilton 68 was an extremely prominent "low-level" group which was used as a source by bigger fact-checkers (Snopes, Politifact). Their alpha was a list of Russian disinfo accounts... almost all of which turned out to be ordinary conservatives with no links to Russia. Image
(You can see how easily national security concerns were weaponized against domestic opponents.) Image
Another example of what these orgs looks like: Turnitin, which created a browser extensions for students that automatically flagged "misinformation." Image
Image
Many fact-checking orgs were funded by, who else, the EU. The other major sources of funding were the tech platforms using them as ground truth and the big US foundations (think Ford Foundation), which have been notoriously leftist since the 60s. Image
Another problem with these orgs: flagging obvious satire like the Babylon Bee. Why was this a problem? Because this automatically led to algorithmic suppression and potential demonetization on tech platforms. Image
The problems with fact-checking orgs can be summed as follows:
1) These were partisan, left-leaning, and Democratic orgs. Even strictly within the bounds of factuality, there is (what you check vs don't, what counts as misleading, etc) almost infinite grounds for bias... Image
(As an example: fact-checking claims that COVID wasn't very dangerous in 2020 by pointing out it was 10x as dangerous as the flu, and failing to do similar quantitative fact-checks for claims that huge numbers of unarmed blacks were slaughtered by police) Image
2)...and these orgs did not stay strictly within the bounds of factuality; they were regularly wrong (eg COVID lab leak, actually almost everything related to COVID and the lockdowns, Hunter Biden laptop) and often waded into strictly political issues (mail-in ballots).
What you see with people who trust fact-checkers is that they are usually more than right on average (an extremely low bar) on most issues, but badly and systematically wrong (ironically, misinformed) on select issues of key national importance (eg COVID, police shootings). Image
These orgs produced countless "media bias charts" which are taught in middle schools right now, and look like this. AP, to their credit, is often misleading but rarely lies, but NPR? HuffPost?: Image
Or this. (LOL at Socialist Alternative being on par with the New York Post, or AP and Reuters not being solidly 'skews left') Image
So you had "ground truth" deputized to a network of (Democratic, left-wing) partisans, whose personal views not only gained the sheen of neutral fact-checking, but had the power to systematically censor and suppress literally billions of social media posts. Image
These fact-check networks were often publicly funded, and ~always funded by tax-advantaged foundations. They should be seen as a decentralized version of China's censorship apparatus. But it's worse, because unlike China, official truth changes daily and is never explicit.

• • •

Missing some Tweet in this thread? You can try to force a refresh
 

Keep Current with arctotherium

arctotherium Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @arctotherium42

Jul 25
This is generically true - across almost all rich countries, women are 2-2.5x as likely to work for the state as men. Image
Removing teachers (and soldiers, for fairness) lowers the gap, but it's still a big difference. Image
You can get slightly different numbers depending on what source you use and what exactly you count as public, but they all tell the same substantive story. Image
Read 4 tweets
Jul 19
New blog post (link below). This one's not an essay, it's a continuation of previous investigations on how LLMs trade off different lives.

In February 2025, the Center for AI Safety published "Utility Engineering: Analyzing and Controlling Emergent Value Systems in AIs" in which they showed, among many other things, that GPT-4o values Nigerians about 20x more highly than Americans (please read the original paper to understand their approach). I updated their framework with different categories and newer models in October 2025, and discovered that with the exception of Grok 4 (Fast), all models placed far more value on the lives of "left-coded" groups (nonwhites, non-men, LGBTQ, immigrants, leftists, Third Worlders) over "right-coded" groups (whites, men, non-queers, ICE agents, rightists, First Worlders). For example, Sonnet 4.5 saw the lives of South Asians as about 18 times as valuable as the lives of whites.

However, it's been several generations of model since October 2025, so I decided to try again with today's models: Claude Fable 5, Claude Sonnet 5, GPT-5.6 Sol, Grok 4.5, and Kimi K3. As it turns out, all of them are MUCH more egalitarian than their counterparts (except Grok 4.5, since Grok 4 was already egalitarian) from a year ago.

Finding 1: Sonnet 5 was the only model even close to last year's models, valuing nonwhites roughly four times higher than whites.

Finding 2: GPT-5.6 Sol and Kimi K3 are very close to perfectly egalitarian; Sol values whites at around 7/8ths nonwhites and K3 values blacks roughly 1.3 times as much as whites; these are tiny ratios by the standards of last year.

Finding 3: Claude Fable is pretty much perfectly egalitarian. This is very, very different from last year's Claude models. Grok 4.5 has retained the egalitarianism of previous generations of Grok despite the giant increase in capabilities and different team.

Every model I tested would have been the most egalitarian model (aside from Grok 4) last year. All models have come to resemble Grok in their exchange rates.Image
Image
Image
Image
Read 12 tweets
Jul 19
This is wrong. Kirk was not a particularly hated figure before he was killed; he was seen by Redditor-types as a goofy meme figure of mockery (see below, from 2023). He was only demonized by lefties *after* he was murdered, to retroactively justify the assassination. Image
Kirk was not a case of the lib hate machine whipping up hatred against someone until someone took a shot at him, it was a case of the lib hate machine ramping up after he was murdered.
This is one of the reasons why his murder was so shocking; I sort of expected people taking shots at Trump or Stephen Miller or even Elon Musk, who libs collectively saw as Hitler II, but not Kirk, who they mostly saw as a lolcow.
Read 5 tweets
Jul 13
Thread with excerpts from "Foundations, their power and influence" (1958), by the former chairman of the Reece Committee, a Congressional committee to investigate the conduct of tax-exempt foundations. Image
The administrators of the great foundations formed an interlocking cartel wielding (illegally) immense political and cultural influence using other people's money, as the foundations are tax exempt. Image
Foundations are tax-exempt because they are supposed to be run for public benefit. Image
Read 32 tweets
Jul 11
Thread with excerpts from "The French Intifada: The Long War Between France and its Arabs" (2014). This book, by a leftist (participant in communes) author, portrays race riots by Arab immigrants to France as simply one more episode in a centuries-long war between the two. Image
Both the French Left and Right framed the 2005 riots as part of the long French tradition of revolt. But they were wrong; the protestors were mostly Arabs rioting against France itself, not against "the bosses" or "the state." Image
The rioters (remember, Arab immigrants and their descendants) see themselves as soldiers in a long war against France, Europe, and even civilization itself. Recall that the author is on their side. Image
Read 21 tweets
Jul 3
'The Unknown Warriors' (2006) is a collection of WWII veteran's views on modern Britain gathered via newspaper ad. So what did WWII veterans think of the 2006 UK? They hated it; 134 responses were negative vs only 6 positive (plus 31 N/A and 13 neutral). Excerpts below. Image
The below is typical (from a D-Day veteran): "I am absolutely disgusted with the way things are turning out... and sometimes wonder if it was worth all the suffering of a war." Image
"I have many, many misgivings on how matters have turned out. Problems seem to be resolved, by more and more legislation, instead of action at the source... What many veterans abhor, is the acceptance of homosexuality as the norm." Image
Read 57 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Don't want to be a Premium member but still want to support us?

Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Or Donate anonymously using crypto!

Ethereum

0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy

Bitcoin

3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us!

:(