SAN FRANCISCO, California, January 12, 2018 (LifeSiteNews) – Conservatives and Christians have long suspected that Twitter works aggressively behind the scenes to censor their material. That suspicion has now been confirmed in an undercover investigation involving numerous Twitter employees who let slip what really happens behind the scenes of the mega social media platform.
A Project Veritas (PV) report released this week zeroes in on Twitter's little-known practice of “shadow banning,” which turns out to be a rather surreptitious, stealth method of censorship.
“Shadow banning is a way of blocking users from a social media platform without notifying them,” explains PV founder, James O’Keefe in the fifteen minute video.
“Tweets from a shadow-banned user still appear to their followers, but don’t show up in search results, or anywhere else on Twitter,” he said.
“Although Twitter presents itself as politically neutral, its culture behind closed doors is one of blatant censorship, systematic bias, and political targeting,” according to O’Keefe.
How 'shadow banning' works
Through shadow banning, “you have ultimate control,” said Abhinav Vadrevu, former Twitter software engineer. “The idea of a shadow ban is that you ban someone but they don’t know they’ve been banned, because they keep posting but no one sees their content. So, they just think that no one is engaging with their content, when, in reality, no one is seeing it.”
“At the end of the day, no one else interacts,” continues Vadrevu. “No one else sees what you’re doing. So, all that data is just thrown away.”
Conrado Miranda, a former Twitter engineer, says shadow banning is happening at Twitter, acknowledging that the procedure is often used to silence conservatives or Trump supporters.
Vadrevu said, “It’s risky though.” Why? “Because people will figure that sh*t out, and be like … you know, it’s a lot of bad press if, like, people figure out like you’re shadow banning them. It’s like, unethical in some way.”
Twitter employs “content review agents,” whose job is to review content that has been flagged as harmful, offensive, or in violation of Twitter policy, and to ban it if it they concur. PV’s O’Keefe describes content review as “an integral part of any social media platform … but it is highly subject to abuse.”
'Shadow banning' criteria
Mo Noral, a former Twitter employee, described to the PV investigative reporters what he did as a content review agent, illustrating the subjective nature of his work.
“Let’s say if it was a pro-Trump thing and I am anti-Trump, I was like, I banned his whole account. It goes to you, and then it’s at your discretion,” he said.
“If this is pro-Trump I don’t want it because it offends me. And I say, 'I banned this whole thing …'”
“On stuff like that, it was more discretion on your viewpoint, I guess how you felt about a particular matter,” continued Noral. He added, “Twitter was probably like 90% anti-Trump, maybe 99% anti-Trump.”
Tweets expressing conservative viewpoints were downranked, while liberal viewpoints were deemed, “OK,” and allowed to continue unchecked, he said.
Twitter’s anti-conservative corporate culture
The PV report indicates that the corporate culture at Twitter is anti-conservative. The report makes the case based on the way certain tweets and Twitter accounts are flagged for review, and then become subject to shadow-banning.
“Just go to a random (Trump) tweet and just look at the followers. They’ll all be like God, ‘Merica, like, and with the American flag, and like, the cross,” says Pranay Singh, a direct messaging engineer for Twitter. He continued, “Who says that? Who talks like that? It’s for sure a bot,” referring to tweets generated by a computer — a “bot” — rather than a real person.
Singh, clearly incredulous, doesn’t believe that people actually love America or embrace Christianity. “There are hundreds of thousands of them, so you’ve got to write algorithms that do it [the censoring] for you.”
Perhaps exaggerating, Singh seemed to say that there are about “5,000 keywords that describe a redneck,” a term Singh used to describe the kind of people that Twitter would like to keep off their platform.
Clay Haynes, a senior network security engineer for Twitter says, “I’m pretty sure every single employee at Twitter hates Trump.”
Olinda Hassan, a Policy Manager for Twitter Trust & Safety, is responsible for Twitter’s rules and regulations, determining what content is and isn’t allowed. Hassan makes clear what Twitter seeks to do: “We’re trying to get the shitty people not to show up. It’s a product thing we’re working on.”
Shadow banning is just one tool Twitter has at its disposal to stealthily target political views it finds offensive and to silence those who express those views.
Twitter's unspoken corporate rules
“[There are] a lot of unwritten rules, and being that we’re in San Francisco, we’re in California, very liberal, a very blue state,” says Noral.
“As a company you can’t really say it because it would make you look bad, but behind closed doors are a lot of rules. Like, hey, you gotta do this this way … it was never written, it was more said.”
PV’s James O’Keefe called Twitters’ unwritten content review protocol a “tool used to promote a political agenda.”
“You won’t find any of these rules or practices outlined in company policy documents,” he added.
O’Keefe said that it “may not be just average users feeling the heat.”
“Does Twitter target certain political figures for suspension? And more importantly, why?”
When asked why certain high-profile political figures or celebrities have their accounts taken down, Pranay Singh, direct messaging engineer, said it might be because of government coercion.
“It might be the U.S. government pressuring us. They don’t like people messing with their politics (and he — Julian Assange — has shit on a lot of people).”
He went on to say that Twitter gets requests from the government to take down celebrities “all the f**king time.”
“What we can do on our side is actually very terrifying,” said senior network engineer Clay Haynes.
Looking to the future: Twitter’s 'secret project'
“Twitter’s censorship practices go beyond a few politically motivated techies,” said O’Keefe. “Twitter is run by an advanced set of algorithms – tools that are used to sift through vast amounts of data, or in this case, ‘tweets.’”
Twitter’s “algorithms block, mute, and prioritize everything on the Twitter platform and [the algorithms] are growing smarter by the day.”
Algorithms can be written which allow machines to do the work done by human censors. For instance, an algorithm can look for certain trigger words, which will cause the platform to automatically delete those tweets and accounts.
In one interchange, Singh suggested that Twitter users with words like “guns” or “Trump” or “conservative” or “America” in their twitter name might be targets for deletion.
Steven Pierre, Twitter Software Engineer, revealed a secret project at Twitter involving “machine learning” i.e., algorithms that teach themselves how to solve problems, eliminating the need for human input.
“Every single conversation is going to be rated by a machine and the machine is going to say whether or not it is a positive thing or a negative thing … it’s going to ban a way of talking,” he said.
“The power over speech Silicon Valley tech giants has is unprecedented and dangerous,” according to Project Veritas’ published report. “What kind of world do we live in where computer engineers are the gatekeepers of the ‘way people talk?’” asked O’Keefe.
“Twitter may fancy itself the global town square, but it’s clear from our reporting that Twitter is not an open, unbiased platform,” concluded O’Keefe.
“Anonymity of its internal policies have bred irresponsibility and abuse. If Twitter wants to convince its users it truly respects free speech, there must be some transparency.”
“Bring shadow banning out of the shadows. Algorithms are only as good as the weights put on them. So, take your thumb off the algorithmic scale. Get rid of the engineers who abuse their power, and show us your HR policies,” he said.